Automatic health and wellness tracking system using machine-learned models

Abstract

Techniques for maintaining a knowledge graph of a user, such as user log, are presented herein. The system can include a database storing the knowledge graph having a plurality of structured data entries. The system can include a machine-learned model configured to determine an insight. The system can obtain, from a user device, a first user entry. Additionally, the system can process, using the machine-learned model, the first user entry with a prompt to generate a first structured data entry. Moreover, the system can process, using the machine-learned model, the first user entry to determine a layout of a graphical user interface that is presented on the user device. Furthermore, the system can process the first structured data entry with the knowledge graph to determine the first insight. Subsequently, the system can cause a presentation of the first insight on the graphical user interface of the user device.

Description

AUTOMATIC HEALTH AND WELLNESS TRACKING SYSTEM USING MACHINE-LEARNED MODELSFIELD

[0001] The present disclosure relates generally to machine learning. More particularly, the present disclosure relates to systems and methods for using machine learning to automatically track and summarize insights generated from user input, structured data, and a user knowledge graph.BACKGROUND

[0002] A health and wellness tracking system, such as a health log, can be a recordkeeping tool used to track various aspects of an individual's health and wellness over time. The purpose of a health log is to help individuals monitor their health-related activities, symptoms, and progress toward their health goals. By maintaining a health log, individuals can gain valuable insights into their health habits, identify patterns or trends, and make informed decisions about their lifestyle, diet, exercise, and healthcare management. However, a conventional system may not be able to accurately maintain a health log without having a user constantly enter log data, which can be time consuming and cumbersome. Current health and wellness tracking systems face a range of technical challenges, such as: integration of data from various devices, sensors, and other mobile applications; ensuring accuracy of data collected; designing mobile applications that are engaging for users to continue using them; handling large volume of data from many users; ensuring low latency in data processing and response; and developing machine-learned models for accurate recommendations and predictions.SUMMARY

[0003] Aspects and advantages of embodiments of the present disclosure will be set forth in part in the following description, or can be learned from the description, or can be learned through practice of the embodiments.

[0004] One example aspect of the present disclosure is directed to an example system for maintaining a knowledge graph of a user. The system can include one or more processors. Additionally, the system can include one or more non-transitory computer-readable media that collectively store a database storing the knowledge graph having a plurality of structured data entries and a machine-learned model. The machine-learned model can be configured to determine an insight. The computer-readable media includes instructions that, when executedby the one or more processors, cause the computing system to perform operations. The operations can include obtaining, from a user device, a first user entry. Additionally, the operations can include processing, using the machine-learned model, the first user entry with a prompt to generate a first structured data entry. Furthermore, the operations can include processing the first structured data entry and the knowledge graph to determine the first insight. Moreover, the operations can include processing, using the machine-learned model, the first insight to determine a layout of a graphical user interface that is presented on the user device. Subsequently, the operations can include causing a presentation of the first insight on the graphical user interface of the user device.

[0005] In some instances, the knowledge graph can be updated by including the first structured data entry in the plurality of structured data entries.

[0006] In some instances, the first user entry can include image data and user speech data.

[0007] In some instances, the first user entry can include video data and user speech data.

[0008] In some instances, the operations further include obtaining an image from the user device. Additionally, the operations can include processing, using an image classification model, the image to determine a classification of an object in the image. The first structured data entry can be determined based on the classification of the image in the model.

[0009] In some instances, the classification of the object can be a food classification. The operations can include obtaining, via the graphical user interface, a user input modifying the classification of the object in the image. Additionally, the operations can include updating the classification of the object in the image based on the second user entry.

[0010] In some instances, the classification is a size associated with the object, and the user input changes the size associated with the object.

[0011] In some instances, the operations further include obtaining, from a nutrition database, nutrition data associated with the classification of the object in the image. The first structured data entry can be generated based on the nutrition data associated with the object.

[0012] In some instances, the operations can include obtaining location data from the user device. The first structured data entry can be generated based on the location data associated with the object. Additionally, the operations can include processing, using the machine-learned model, the first structured data, the knowledge graph, and the location data to determine the first insight.

[0013] In some instances, the operations can include determining the layout of the graphical user interface based on an attribute of a display screen of the user device.

[0014] In some instances, the operations can include determining the layout of the graphical user interface based on the first insight.

[0015] In some instances, the operations can include determining the layout of the graphical user interface based on an attribute of the first structured data entry.

[0016] In some instances, the operations can include analyzing the knowledge graph to determine a patterned-input received at a specific time. Additionally, the operations can include prompting for a second user entry after determining that the patterned-input has not been received by the specific time.

[0017] In some instances, the operations can include obtaining, from a sensor device, sensor data associated with a user of the user device. Additionally, the operations can include processing, using the machine-learned model, the sensor data to generate a second structured data entry. Furthermore, the operations can include storing the second structured data entry in the knowledge graph.

[0018] In some instances, the operations can include obtaining, from a sleep pattern database, a first sleep pattern associated with the sensor data. The first structured data entry can be generated based on the first sleep pattern.

[0019] In some instances, the insight is a new habit of the user, and the machine-learned model tracks the new habit of the user.

[0020] Another example aspect of the present disclosure is directed to an example computer-implemented method for maintaining a knowledge graph having a plurality of structured data entries. The method can include obtaining, from a user device, a first user entry. Additionally, the method can include processing, using a machine-learned model, the first user entry with a prompt to generate a first structured data entry. Moreover, the method can include updating the knowledge graph by including the first structured data entry in the plurality of structured data entries. Furthermore, the method can include processing, using the machine- learned model, the first user entry to determine a layout of a graphical user interface that is presented on the user device. The method can include processing the first structured data entry with the knowledge graph to determine the first insight. Subsequently, the method can include causing a presentation of the first insight on the graphical user interface of the user device.

[0021] In some instances, the method can include obtaining an image from the user device. Additionally, the method can include processing, using an image classification model, the image to determine a classification of an object in the image. The first structured data entry can be further determined based on the classification of the image in the model.

[0022] In some instances, the method can include analyzing the knowledge graph to determine a patterned-input received at a specific time. Additionally, the method can include prompting for a second user entry after determining that the patterned-input has not been received by the specific time.

[0023] Another example aspect of the present disclosure is directed to one or more non- transitory computer-readable media. The one or more non-transitory computer-readable media can store instructions that are executable by a computing system to perform example operations. The operations can include obtaining, from a user device, a first user entry. Additionally, the operations can include processing, using the machine-learned model, the first user entry with a prompt to generate a first structured data entry. Moreover, the operations can include processing, using the machine-learned model, the first user entry to determine a layout of a graphical user interface that is presented on the user device. Furthermore, the operations can include processing the first structured data entry with the knowledge graph to determine the first insight. Subsequently, the operations can include causing a presentation of the first insight on the graphical user interface of the user device.

[0024] Other aspects of the present disclosure are directed to various systems, apparatuses, non-transitory computer-readable media, user interfaces, and electronic devices.

[0025] These and other features, aspects, and advantages of various embodiments of the present disclosure will become better understood with reference to the following description and appended claims. The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate example embodiments of the present disclosure and, together with the description, serve to explain the related principles.BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Detailed discussion of embodiments directed to one of ordinary skill in the art is set forth in the specification, which makes reference to the appended figures, in which:

[0027] Figure 1 depicts a block diagram of an example system for generating structured data entry to be stored in a knowledge graph according to example embodiments of the present disclosure.

[0028] Figure 2 depicts a block diagram of an example system for processing user input using a machine-learned model to generate an insight according to example embodiments of the present disclosure.

[0029] Figure 3 depicts a flow chart diagram of an example method to generate insight according to example embodiments of the present disclosure

[0030] Figure 4 depicts graphical user interfaces of an example system for obtaining user entries according to example embodiments of the present disclosure.

[0031] Figure 5 depicts a graphical user interface of an example system for an interactive user experience according to example embodiments of the present disclosure.

[0032] Figure 6 depicts a graphical user interface of an example system for presenting an insight according to example embodiments of the present disclosure.

[0033] Figure 7 A depicts a block diagram of an example computing system that performs insight summary generation according to example embodiments of the present disclosure.

[0034] Figure 7B depicts a block diagram of an example computing device that performs insight summary generation according to example embodiments of the present disclosure.

[0035] Figure 7C depicts a block diagram of an example computing device that performs insight summary generation according to example embodiments of the present disclosure.

[0036] Figure 8 depicts a flowchart of a method for training one or more machine-learned models according to aspects of the present disclosure.

[0037] Figure 9 is a block diagram of an example processing flow for using machine- learned model(s) 1 to process input(s) 2 to generate output(s) 3.

[0038] Figure 10 is a block diagram of an example implementation of an example machine-learned model configured to process sequences of information.

[0039] Figure 11 is a block diagram of an example technique for populating an example input sequence.

[0040] Figure 12 is a block diagram of an example model development platform that can facilitate creation, adaptation, and refinement of example machine-learned models.

[0041] Figure 13 is a block diagram of an example training flow for training a machine- learned development model.

[0042] Figure 14 is a block diagram of an inference system for operating one or more machine-learned model(s) to perform inference.

[0043] Reference numerals that are repeated across plural figures are intended to identify the same features in various implementations.DETAILED DESCRIPTION

[0044] Generally, the present disclosure is directed to machine-learned generation of insight summaries based on health and fitness data. A computing system can perform an analysis of the health and fitness data to generate insights about the health and fitness of a user. For example, the analysis can identify particular trends in the data over time, such as a recentincrease or decrease in a numerical value associated with the health and fitness data. The system can provide insights that are generated by a generative machine-learned model (e.g., generative language model), and the model can generate a summary of the insight data in a natural language. The insight summary can then be provided to a user to help the user better understand aspects of the health and fitness data.

[0045] Capturing a comprehensive understanding of the health and wellness of a user typically requires manual logging of data by the user. Thus, in a conventional system, the user needs to manually input the subjective data and behavioral data. For example, a health provider may ask users to keep a journal and manually log their health and wellness information. In contrast, the system described herein automates the process to collect the relevant information and generates a knowledge graph based on the relevant information. The relevant information can be derived by processing multimodal input data (e.g., natural language, photo, video) using a large language model (LLM). The relevant information can be stored as structured input data in a knowledge graph for tracking progress toward health and fitness goals.

[0046] According to embodiments of the present invention, a health and wellness tracking system, using machine-learned models, can have the capabilities to solve the shortcomings of manually entering logging data and enable analysis of data to provide insights to a user. For example, using the system described herein, the machine-learned models, such as large language models (LLMs), can automatically perform the functions of logging and can provide a better user experience with an improved user interface. The system can provide an artificial intelligence (Al) powered experience that uses LLMs for quick input and personal insights. In some instances, the interaction time with the system described herein can be 25 times faster than a conventional mobile application system. Typical logging which traditionally would take minutes can now be performed in seconds. Additionally, the LLM can be multi-modal and can process user input (e.g., text, voice, photos, and videos) in real-time with context data related to the user. The context data can be derived from a knowledge graph of the user. The knowledge graph can be a user log of the health and wellness tracking system. Additionally, the context data include sensor data from a sensor associated with a user. In some instances, the context data can include user data obtained from a remote server. The user data can include location data, personal preferences, and personal goals.

[0047] In some instances, the system can log data by allowing users to speak naturally in short or long form journal entries. Health and wellness data can be determined (e.g., derived) from the conversation and the LLM can generate insights, which can include a daily overview of the user’s activities. The system is designed to be flexible to enable a user to log any type ofinformation. With natural language input and LLM capabilities, users can log information that may not be possible with a conventional logging application. The system can merge food logging, mood logging, journaling, and other user information.

[0048] The system can determine insights by combining logs with sensor data. The insights can be about personal health and wellness. Additionally, the system can track new habits. For example, the system can identify healthy habits to focus on, and enable users to create, track and share these habits.

[0049] Further to the descriptions above, a user may be provided with controls allowing the user to make an election as to both if and when systems, programs, or features described herein may enable collection of user information (e.g., sensor data, log data, information about user activities, user preferences, user location, or other user data), and if the user is sent content or communications from a server. In addition, certain data may be treated in one or more ways before it is stored or used, so that personally identifiable information is removed. For example, a users identity may be treated so that no personally identifiable information can be determined for the user, or a user’s geographic location may be generalized where location information is obtained (such as to a city, ZIP code, or state level), so that a particular location of a user cannot be determined. Thus, the user may have control over what information is collected about the user, how that information is used, and what information is provided to the user.

[0050] In particular, in some implementations, the computing system can remove personal information from the obtained data so that information such as specific user identity or user identifier is not included in the data. For example, in some implementations, the computing system can respectively assign a unique and non-personally identifying identifier to the sensor data and log data for each of the first plurality of users and then remove any personal information from the sensor data and log data. As examples, the unique and non-personally identifying identifier can include a random number, a result of a hash function applied to a user identifier, or some other obfuscated or encrypted identifier which cannot be reverted to obtain the user identifier. Likewise, the computing system can respectively assign the same unique and non-personally identifying identifier to the location data for each of the first plurality of users and then remove personal information from the location data.

[0051] In some instances, the knowledge graph, which includes structured data entries and insights that have been derived from user information (e.g., user input data and sensor data), can be encrypted and stored on a local computing device of the user (e.g., mobile device of the user). As such, in some instances, the sensitive information is stored on a local device of the user and not shared with a client server. In another example, based on user election andpreference, certain data may be stored in a client server treated in one or more ways before it is stored or used, so that personally identifiable information is removed. As an example, the unique and non-personally identifying identifier can include a random number, a result of a hash function applied to a user identifier, or some other obfuscated or encrypted identifier which cannot be reverted to obtain the user identifier. As previously mentioned, the user may have control over what information is collected about the user and how that information is used.

[0052] According to some embodiments, a user entry (e.g., user input, log entry), can be multimodal. The user input can be text, audio, image, and / or video data. The audio data can be converted to text data. The system can determine an object in the image data using image embeddings. By determining objects in the image data, the system can convert the image data into text data (e.g., object names). Similarly, the video data can be converted to text data, by converting the video data into a plurality of images. Subsequently, the text data and a prompt can be inputted into the machine-learned model. The prompt can be generated based on the user input. In some instances, the prompt is selected from a plurality of predetermined prompts based on the user input and / or knowledge graph of the user.

[0053] In some instances, example systems can use various methods to reduce a risk of error associated with language generation. For example, some machine-learned language generation models may not include a mathematical analysis component or a fact-checking component, and thus may in some instances generate mathematically or factually erroneous outputs if no error prevention methods are employed. Example embodiments of the present disclosure can include various techniques for error prevention, detection, and correction.

[0054] For example, in some instances, the analysis can include a deterministic analysis (e.g., mathematical analysis, algorithmic analysis, etc.) that is guaranteed to process data in a mathematically accurate way. In this manner, for instance, the generative machine-learned model can base its summaries on insight data that has already been processed mathematically and is guaranteed to be accurate when provided in a structured format to the generative machine-learned model.

[0055] As another example, in some instances, the generative machine-learned model can generate multiple candidate summaries associated with multiple potential insights, and the candidate summaries can be evaluated by one or more separate machine-learned evaluation models. For example, a separate evaluation model can be separately trained to detect whether mathematical or factual claims in a natural language output are supported by the input data that the generated sequence is based on. The evaluation model(s) can also evaluate candidateoutputs to detect whether the outputs comply with other goals, such as compliance with formatting instructions, readability, actionability, and the like. In some instances, a computing system can select, based on the evaluations, one best output from the candidate outputs to display to a user. In some instances, an evaluation threshold (e.g., accuracy confidence threshold, readability score threshold) can be used, and the computing system can decide not to show any insight summaries to a user if none of the candidate outputs meet the threshold. In this manner, for instance, a computing system can ensure that any machine-learned output provided to a user will be accurate and useful.

[0056] As another example, the output generation process can include various additional guardrails to reduce a risk of generating flawed (e.g., mathematically flawed, improperly formatted, too long or short) candidate outputs. For example, an input to the generative machine-learned model can include a fill-in-the-blank-style template, along with an instruction to fill in the blanks based on the structured input data. In this manner, for instance, the range of possible machine-learned outputs can be narrowed to a range that is likely to generate high- quality outputs (e.g., likely to comply with formatting goals and readability goals, unlikely to result in mathematically erroneous outputs). As another example, an input to the generative machine-learned model can include multiple input-output pairs that include a structured insight data input and a high-quality insight summarization output. In this manner, for instance, example embodiments can capitalize on the in-context learning capabilities of some generative machine-learned models to improve the quality of generated candidate outputs. As another example, an input to the generative machine-learned model can include general knowledge (e.g., retrieved factual knowledge, general advertising analytics knowledge) that may reduce an error rate or otherwise improve the candidate outputs by providing relevant context to capitalize on the in-context learning capabilities of some generative machine-learned models.

[0057] As another example, the insight generation process can be iteratively improved based on feedback from users. For example, a system can provide a generated insight to a user, along with an input component for the user to provide feedback about the quality (e.g., accuracy, relevance, interestingness, usefulness, actionability) of the generated insight. Based on feedback received via the input component, a computing system can further train the evaluation model, the generative machine-learned model, or both to further improve the quality of generated outputs.

[0058] In some implementations, the techniques disclosed herein enable artificial intelligence to generate accurate and relevant insights. Artificial intelligence (Al) is a segment of computer science that focuses on the creation of models that can perform tasks with little tono human intervention. Artificial intelligence systems can utilize, for example, machine learning, natural language processing, and computer vision. Machine learning, and its subsets, such as deep learning, focus on developing models that can infer outputs from data. The outputs can include, for example, predictions and / or classifications. Natural language processing focuses on analyzing and generating human language. Computer vision focuses on analyzing and interpreting images and videos. Artificial intelligence systems can include generative models that generate new content, such as images, videos, text, audio, and / or other content, in response to input prompts and / or based on other information.

[0059] Systems and methods of the present disclosure can provide a variety of technical effects and benefits, such as improved accuracy of machine-learned outputs; reduced computational cost (e.g., electricity cost, processor usage, etc.) of machine-learned language generation; and reduced cost (e.g., computational cost, labor cost, etc.) of insight extraction. In some instances, techniques described herein can automatically generate insights, structured data entries, and a knowledge graph to obtain a comprehensive understanding of the health and wellness of a user without manual logging of data by the user.

[0060] For example, systems and methods according to example aspects of the present disclosure can provide improved accuracy of machine-learned outputs. Some alternative methods may provide raw or unstructured data (e.g., picture of a meal, natural language audio input) to a machine-learned language generation model, which may cause the language generation model to generate language outputs that include factually incorrect assertions. As another example, some alternative methods may be configured to provide machine-generated outputs to a user without a mechanism to evaluate the outputs’ accuracy or to filter out inaccurate outputs. Advantageously, systems and methods according to the present disclosure can reduce a rate of factual errors in outputs generated by a machine-learned generative language model itself, and in outputs provided to the user by a computing system comprising the machine-learned generative language model, compared to alternative methods with fewer or less effective error prevention mechanisms.

[0061] As another example, systems and methods according to example aspects of the present disclosure may in some instances reduce a computational cost of generating machine- learned insight outputs compared to some alternative methods with a similar accuracy. For example, in some instances, the factual accuracy of a machine-learned language output can be increased by increasing the complexity or size (e.g., number of parameters) of the machine- learned model generating the output. However, increasing a complexity of a machine-learned model can also increase a computational cost (e.g., electricity cost, processor usage, memoryusage, hardware cost) of training the machine-learned model and a computational cost of generating outputs with the machine-learned model after training. In some instances, the increased cost can be very large compared to the improvement in accuracy. For example, a large increase in model complexity (e.g., doubling of parameter count, ten-fold increase in parameter count) may only lead to a small marginal increase in accuracy in a simple mathematical reasoning task, which may be much simpler mathematically than structured data analysis performed according to some aspects of the present disclosure. Additionally, the increase in accuracy may in some instances have a log-linear relationship with model complexity, meaning that increased complexity will lead to diminishing returns in accuracy as model complexity increases. Advantageously, systems and methods according to some aspects of the present disclosure can provide substantially improved accuracy compared to alternative methods, without increasing the complexity of the machine-learned language model. In this manner, for instance, systems and methods according to some aspects of the present disclosure can provide machine-learned insight summarization at reduced computational cost (e.g., model training costs, inference costs) compared to alternative methods having a similar accuracy.

[0062] A technical effect of example implementations of the present disclosure is increased energy efficiency in performing operations using machine-learned models, thereby improving the functioning of computers implementing such models. For instance, example implementations can provide for more energy-efficient training operations or model updates by providing error correction using lightweight (e.g., having a lower computational cost or model complexity compared to a machine-learned generative language model) evaluation models or structured data analysis techniques. In some scenarios, increased energy efficiency can provide for less energy to be used to perform a given number of inference or training tasks (e.g., less energy expended to maintain the model in memory, less energy expended to perform calculations within the model, such as computing gradients, backpropagating a loss). In some scenarios, increased energy efficiency can provide for more inference or training tasks to be completed for a given energy budget (e.g., a larger quantity of training iterations). In some scenarios, greater expressivity afforded by systems and methods of the present disclosure can provide for a given level of functionality to be obtained in fewer training iterations, thereby expending a smaller energy budget. In some scenarios, greater expressivity afforded by systems and methods of the present disclosure can provide for an extended level of functionality to be obtained in a given number of training iterations, thereby more efficiently using a given energy budget.

[0063] Additionally, by automatically generating structured data entry that is then stored in a master knowledge graph, the system can ensure the accuracy of data collected from various sources, including user input and wearable devices. Additionally, the custom layout of the graphical user interface (e.g., based on the determined insight) can enable the system to engage and motivate users’ continuous use. Additionally, the system, using the knowledge graph, can personalize the user experience to cater to individual needs and preferences. Furthermore, the structured data entry that is stored in the knowledge graph ensures interoperability between different health information systems, and standardizes data formats for seamless data exchange. The structured data entries also allow for scalability, which enables the system to handle large volumes of data from a growing number of users, and ensures that the system can scale efficiently without compromising performance. The structured data entries and knowledge graph ensures low latency in data processing, which enables the system to provide real-time insights (e.g., analysis) and feedback to users based on data from wearable devices, user entries and other inputs. The graphical user interface that presents the Al-generated insights and the structured data entries ensuring the transparency and explainability of Al-generated insights.

[0064] With reference now to the Figures, example embodiments of the present disclosure will be discussed in further detail.Example Systems

[0065] Figure 1 depicts a block diagram of an example system 100 for generating a structured data entry. In some instances, the system 100 can be, comprise, be comprised by, or share one or more properties with a computing device or system described below with respect to Figures 7A-7C (e.g., server computing system 730, training computing system 750, computing device 10, computing device 50).

[0066] According to some embodiments, the system 100 can receive a user entry 110, selecting a prompt 120, and feeding the user entry and prompt into the LLM to generate the structured data entry 130 structured data (e.g., XML, JSON, HTML, object, struct. For example, the system can feed the user entry 110 into the LLM with a prompt 120 that returns structured entry 130. The user entry 110, which is a journal entry in this example, can state “for breakfast I had some oatmeal with fruit, two coffees and a small glass of orange juice. I woke up feeling pretty good and ready for the day.” The prompt 120, which can be selected from a plurality of predetermined prompts, can state “please return a valid JSON object that lists which words are food. ..” In this example, the structured entry 130 can be a JSON object with time, categorization, mood, portions, and activities. Additionally, the knowledge graph 140 (e.g.,user log) of the user can be updated to include the generated structured data entry as a node. This generated structured data entry can be connected to other nodes based on a common attribute. The system 100 can perform a categorization of user input, highlight key terms in a user input, provide weekly summaries, and enable habit tracking.

[0067] A structured data entry 106 can include, for example, one or more data items in a structured format. In some instances, the structured data entry 106 can include data items correlating numerical data derived from aggregated statistical data (e.g., trends). As an illustrative example, structured data entry 106 associated with a lunch image of a user includes a time stamp, object names of the items in the lunch image, nutritional data for each item, and so on.

[0068] A structured data entry 106 can include, for example, data objects (e.g., associated with an object-oriented programming language) or data structures (e.g., structs in a C programming language); database rows or spreadsheet rows; data in a structured text format, such as a data object notation format (e.g., Javascript Object Notation (JSON) format), markup language format (e.g., extensible markup language (XML) format, hypertext markup language (HTML) format), or other structured format (e.g., comma-separated value (CSV) format, etc.); ordered tuplets or other data formatted according to a predefined order or arrangement; structured format associated with a communication protocol or data storage protocol; files comprising data in a structured format; or other structured data.

[0069] A structured data entry 106 can include one type or many types of data. Example data types for the structured data 106 can include any type of computer-readable data, such as numerical data, binary data, text data, structured data (e.g., XML, JSON, HTML), or other computer-readable data type.

[0070] In some instances, a structured data entry 106 can be derived based on a trend detection operation that compares an attribute (e.g., daily steps taken, hours slept) associated with a first plurality of times and the attribute associated with a second plurality of times. As a non-limiting illustrative example, the trend detection operation can compare, for each of a plurality of attributes, a first plurality of values of the attribute over a first plurality of times (e.g., recent time period such as the past 24 hours, past 48 hours, past week) to a second plurality of values of the attribute over a second plurality of times (e.g., less recent time period such as prior day, week, month, or year before the recent time period began; lifetime of a user account or other advertising analytics account, e.g., at all times before the recent time period began). In some instances, a structured analysis can include a comparison between a first trend associated with a first time period (e.g., most recent 24 hours, etc.) and a second trendassociated with a second time period (e.g., same date or holiday one year earlier; same day of the week or month, one week or month earlier) or plurality of second time periods (e.g., average trend on same day every week, month, year).

[0071] In some instances, a comparison between a first trend and a second trend can include a benchmarking operation, wherein the first trend associated with a user or account of interest (e.g., user to whom the output will be provided) can be compared to second trend associated with a plurality of users or accounts.

[0072] The knowledge graph 140 can include a plurality of structured data entries associated with daily activities, symptom tracking, medication, treatment, vital signs, health measurements, health goals, and so on. The daily activities can include details about physical activities, exercise routines, dietary intake, hydration, and sleep patterns. The symptom tracking can include structured data entries that record user symptoms, such as pain, fatigue, headaches, digestive issues, or changes in mood. The medication and treatment information may include a record of medications taken, including dosage and frequency, as well as any other treatments or therapies undergone. The vital signs can include blood pressure, heart rate, temperature, and weight. The health measurements can include health metrics such as blood sugar levels, cholesterol levels, or other relevant biomarkers. Additionally, a user can set specific health goals, such as weight loss, improved fitness, better stress management, or better management of a chronic condition, and the system 100 can track progress toward these goals over time.

[0073] The system 100 can provide features and functionalities to make the logging process more convenient, intuitive, and effective for users. The system 100 enables a user to create a profile by inputting basic information such as age, gender, height, weight, and any relevant health conditions or goals. Additionally, the system 100 includes an interface that provides a user-friendly layout for inputting daily health-related activities, symptoms, and other data points. Users can easily input information. The user entry 110 can include physical activities and exercises performed, including duration and intensity. Additionally, the user entry 110 can include dietary intake, such as meals, snacks, and beverages consumed. Moreover, the user entry 110 can include hydration levels, water intake, and other fluids consumed. Furthermore, the user entry 110 can include sleep patterns, including bedtime, wake-up time, and sleep quality. The user entry 110 can include health information such as symptoms experienced, medications taken, vital signs (e.g., obtained from a sensor on a wearable device, and health measurements (e.g., blood pressure, heart rate, weight, and temperature).

[0074] The system 100 can process the user entry 110 with a selected prompt 120 to generate a structured data entry 130. The structured data entry 130 can be labeled with one or more attributes. Based on the attributes, the system can generate a customized tracking of the categories determined to be relevant for a user based on the knowledge graph 140. For example, the system can track blood sugar levels for a user managing diabetes, while the system can track workouts and calorie intake for a fitness enthusiast.

[0075] Additionally, the system 100 provides customized user interface, visualizations, and charts based on the knowledge graph 140 to help users analyze their data over time. The system, using the machine-learned models, can present trends, patterns, correlations between different variables (e.g., based on attributes), and progress toward health goals. Users can easily view their data in daily, weekly, monthly, or custom time intervals.

[0076] Moreover, a user can set specific health goals (e.g., achieving a target weight, increasing daily steps, or lowering blood pressure), and the system can automatically track progress toward these goals, generate reports, and provide feedback and reminders to the user.

[0077] Furthermore, the system obtains sensor data that is obtained from a wearable device. For example, the system can receive sensor data (e.g., activity level, heart rate, body temperature, sleep pattern) from a wearable device (e.g., fitness tracker, smartwatch, and health monitoring device). The system can process the sensor data to generate a structured data entry 130.

[0078] The system converts user input into a knowledge graph which is displayed as dynamically generated UI components to assist the user in adding or editing data and confirming the data is entered correctly. The knowledge graph 140 enables health and wellness tracking that is customized for each user based on target goals of each user.Example Model Arrangements

[0079] Figure 2 depicts a block diagram of a system 200 according to example embodiments of the present disclosure. The system 200 can include a user device 205 (e.g., smartphone) having a health app 206, a virtual assistant device 210 (e.g., smart speaker), biometric data 215, speech-to-text data 220, preference data 225, a machine-learned model 235, structured data 240, knowledge graph 245, wearable device data 250, and a user device 255 with a graphical user interface 260.

[0080] The components 205-260 can communicate with each other and work together to collect, process, store, and serve targeted content to users efficiently. By orchestrating components 205-260 effectively, the system 500 can automatically generate a health and fitnessdatabase (e.g., log) for a user in a convenient and user-friendly manner, while minimizing operational overhead.

[0081] The user device 205 can transmit a plurality of user entries to the system 200. The user entry can be a multimodal input (e.g., image data, audio data) that is obtained from various components (e.g., display, microphone, camera) of the user device 205.

[0082] Similarly, the virtual assistant device 210 can transmit a plurality of user entries to the system 200. The user entry can be audio data that is obtained from a microphone of the virtual assistant device 210.

[0083] In some embodiments, the user device 205 and / or virtual assistant device 210 can provide user credentials to the system to authenticate the user. For example, certain platforms may require login credentials or other security credentials before allowing access to the knowledge graph 245 of the user. The system 200 can obtain the required credentials from a user or from a credential storage location.

[0084] The system can include biometric data 215 from a wearable device. The biometric data can include activity data, heart rate data, sleep data, exercise data, health data, and location data. The activity data can include: number of steps taken each day, the distance covered while walking or running, the number of floors or elevation changes, and the amount of time spent in physical activity. The heart rate data can include resting heart rate, active heart rate, and heart rate variability. The sleep data can include sleep duration, sleep stages (e.g., light, deep, REM). The exercise data can include exercise type, exercise duration, calories burned, and so on. The health data can include calories consumed, weight, hydration levels, stress levels, oxygen saturation, and skin temperature. The location data can include GPS information.

[0085] The system can include user data 230. The user data 230 can include a broad range of information related to an individual's health and fitness. The user data 230 can include activity and fitness data, health metrics, nutrition data, lifestyle data, and habits data. The system 200 can provide personalized insights, recommendations, and coaching to help users achieve their health and fitness goals based on the user data 230. The user data 230 can also be used for tracking progress over time, setting and achieving goals, and improving overall wellbeing.

[0086] In some embodiments, the system 200 can perform data conversion on the received user entry to a format usable by the machine-learned model 235. For example, the system 200 can include a natural language process to convert speech-to-text input 220. For example, machine-learned models 235 can process data in different reporting formats, data formats, and the like.

[0087] The machine-learned model 235 can include a large language model. FIGS. 7A-14 further describes the machine-learned model 235.

[0088] The machine-learned model 235 can generate structured data 240 which can be used to update the knowledge graph 245 of the user as described in FIGS. 1 and 3. Additionally, the knowledge graph 245 can also include wearable device user data 250.

[0089] The system 200 can present the insights on a graphical user interface 260 of the user device 255. The insights can be presented to a user in real-time. In some instances, the system 200 can interface with websites, mobile apps, or other digital platforms based on the insight.

[0090] Further to the descriptions above, a user may be provided with controls allowing the user to make an election as to both if and when systems, programs, or features described herein may enable collection of user information (e.g., information about a user’s social network, social actions, or activities, profession, a user’s preferences, or a user’s current location), and if the user is sent content or communications from a server. In addition, certain data may be treated in one or more ways before it is stored or used, so that personally identifiable information is removed. For example, a user’s identity may be treated so that no personally identifiable information can be determined for the user, or a user’s geographic location may be generalized where location information is obtained (such as to a city, ZIP code, or state level), so that a particular location of a user cannot be determined. Thus, the user may have control over what information is collected about the user, how that information is used, and what information is provided to the user.Example Methods

[0091] Figure 3 depicts a flow chart diagram of an example method for maintaining a knowledge graph of a user according to example embodiments of the present disclosure. Although Figure 3 depicts steps performed in a particular order for purposes of illustration and discussion, the methods of the present disclosure are not limited to the particularly illustrated order or arrangement. The various steps of method 300 can be omitted, rearranged, combined, and / or adapted in various ways without deviating from the scope of the present disclosure.

[0092] The system can include a database storing the knowledge graph. The knowledge graph can have a plurality of structured data entries. Additionally, the system can include a machine-learned model. The machine-learned model can be configured to determine an insight.

[0093] At 302, the system can obtain, from a user device, a first user entry (e.g., first user input).

[0094] In one embodiment, the first user entry can include image data and user speech data. In another embodiment, the first user entry can include video data and user speech data.

[0095] At 304, the system can process, using the machine-learned model, the first user entry with a prompt to generate a first structured data entry.

[0096] In some instances, the system can obtain an image from the user device. For example, the first user entry obtained at 302 can include image data. The image data can include an image. Additionally, the system can process, using an image classification model, the image to determine a classification of an object in the image. Moreover, the first structured data entry that is determined at 304 can be based on the classification of the image in the model.

[0097] According to one embodiment, the classification of the object in the image data can be a food classification. The system can further obtain, via the graphical user interface, a second user input modifying the classification of the object in the image. Additionally, the system can update the classification of the object in the image based on the second user entry. In one example, the classification can be a size associated with the object, and the user input can change the size associated with the object. In this example, the user input can indicate the size of the food (e.g., large fries). In some instances, the classification, which may include nutritional information and / or size information, of the object can be updated to change the calorie count of the object and / or size of the object

[0098] In some instances, the system can obtain, from a nutrition database, nutrition data associated with the classification of the object in the image. Additionally, the first structured data entry that is generated at 304 can be based on the nutrition data associated with the object.

[0099] In some instances, the system can obtain, from a sleep pattern database, a first sleep pattern associated with the sensor data. Additionally, the first structured data entry that is generated at 304 can be based on the first sleep pattern.

[0100] At 306, the system can process the first structured data entry and the knowledge graph to determine the first insight.

[0101] In some instances, the insight is a new habit of the user, and the machine-learned model tracks the new habit of the user. For example, the new habit can be a physical goal such as a goal of walking 10,000 steps on a daily basis.

[0102] In some instances, the system can obtain location data from the user device. Additionally, the first structured data entry that is generated at 304 can be based on the location data associated with the object. Furthermore, the system can process, using the machine- learned model, the first structured data, the knowledge graph, and the location data to determine the first insight at 308.

[0103] At 308, the system can process, using the machine-learned model, the first insight to determine a layout of a graphical user interface that is presented on the user device.

[0104] In some instances, the system can determine the layout of the graphical user interface based on an attribute of a display screen of the user device. For example, the layout can be determined based on the resolution or size of the display screen.

[0105] In some instances, the system can determine the layout of the graphical user interface based on an attribute of the insight. For example, the first insight can be nutritional information, and the layout can be determined based on the best way to present the nutritional information about the meal. In one example, the layout can include trend information or total daily calorie count.

[0106] In some instances, the system can determine the layout of the graphical user interface further based on the first insight that is determined at 308. For example, the first insight can be suggested fitness goal (e.g., 10,000 steps per day), and the layout can be based on that specific suggested fitness goal.

[0107] At 310, the system can cause a presentation of the first insight on the graphical user interface of the user device. The graphical user interface can have the layout that is determined at 306.

[0108] Additionally, the knowledge graph can be updated by including the first structured data entry that is generated at 304 in the plurality of structured data entries.

[0109] In some instances, the system can analyze the knowledge graph to determine a patterned-input received at a specific time. Additionally, the system can prompt for a second user entry after determining that the patterned-input has not been received by the specific time. For example, the system typically received a user entry from a user at a specific time of the day (e.g., dinner time), and based on that pattern, the system can prompt the user to a user entry (e.g., input related to user’s dinner) when the system determines that the user entry has not been received with the estimated time limit.

[0110] In some instances, the system can obtain, from a sensor device, sensor data associated with a user of the user device. Additionally, the system can process, using the machine-learned model, the sensor data to generate a second structured data entry. Furthermore, the system can store the second structured data entry in the knowledge graph. For example, the first insight determined at 308 can be associated with the number of daily steps, and the system can obtain step information (e.g., number of steps) from awearable device (e.g., smartwatch) of the user. The step information can be stored as a second structured data entry in the knowledge graph.

[0111] Figure 4 depicts graphical user interfaces of an example system for obtaining user entries according to example embodiments of the present disclosure. For example, the first graphical user interface 400 can be customized based on a user entry (e.g., user request) for a weekly summary. In another example, the user entry can be an inquiry to the system to determine the cause of an ailment (e.g., waking up tired). In this example, the second graphical user interface 410 can provide an analysis based on the knowledge graph of the user. In yet another example, the third graphical user interface 420 can be generated based on a user entry for tracking a specific habit. The third graphical user interface 420 can provide a list of items that is currently being tracked by the system.

[0112] Figure 5 depicts a graphical user interface of an example system for an interactive user experience according to example embodiments of the present disclosure. In the graphical user interface 500, the user can select an item in the picture and provide additional input. For example, the user can select the first item (e.g., bread) and provide additional input (e.g., indicating that it was whole wheat bread).

[0113] Figure 6 depicts a graphical user interface of an example system for presenting an insight according to example embodiments of the present disclosure. In the graphical user interface 600, the system can present insights that are generated based on the knowledge graph.Example Devices and Systems

[0114] Figure 7A depicts a block diagram of an example computing system 700 that performs insight summary generation according to example embodiments of the present disclosure. The system 700 includes a user computing device 702, a server computing system 730, and a training computing system 750 that are communicatively coupled over a network 780.

[0115] The user computing device 702 can be any type of computing device, such as, for example, a personal computing device (e.g., laptop or desktop), a mobile computing device (e.g., smartphone or tablet), a gaming console or controller, a wearable computing device, an embedded computing device, or any other type of computing device.

[0116] The user computing devi ce 702 includes one or more processors 712 and a memory714. The one or more processors 712 can be any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, an FPGA, a controller, a microcontroller, etc.) and can be one processor or a plurality of processors that are operatively connected. The memory 714 can include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, flash memory devices, magnetic disks, etc., and combinations thereof.The memory 714 can store data 716 and instructions 718 which are executed by the processor 712 to cause the user computing device 702 to perform operations.

[0117] In some implementations, the user computing device 702 can store or include one or more machine-learned models 720, such as machine-learned generation models 108 or machine-learned evaluation models 112. For example, the machine-learned models 720 can be or can otherwise include various machine-learned models such as neural networks (e.g., deep neural networks) or other types of machine-learned models, including non-linear models and / or linear models. Neural networks can include feed-forward neural networks, recurrent neural networks (e.g., long short-term memory recurrent neural networks), convolutional neural networks or other forms of neural networks. Some example machine-learned models can leverage an attention mechanism such as self-attention. For example, some example machine- learned models can include multi-headed self-attention models (e.g., transformer models). Example machine-learned models 720 are discussed with reference to Figures 1-3.

[0118] In some implementations, the one or more machine-learned models 720 can be received from the server computing system 730 over network 780, stored in the user computing device memory 714, and then used or otherwise implemented by the one or more processors 712. In some implementations, the user computing device 702 can implement multiple parallel instances of a single machine-learned model 720 (e.g., to perform parallel insight summary generation or evaluation across multiple instances of machine-learned generation model 108 or machine-learned evaluation model 112).

[0119] Additionally or alternatively, one or more machine-learned models 740 (e.g., machine-learned generation models 108, machine-learned evaluation models 112, etc.) can be included in or otherwise stored and implemented by the server computing system 730 that communicates with the user computing device 702 according to a client-server relationship. For example, the machine-learned models 740 can be implemented by the server computing system 730 as a portion of a web service (e.g., an advertising analytics service, etc.). Thus, one or more models 720 can be stored and implemented at the user computing device 702 and / or one or more models 740 can be stored and implemented at the server computing system 730.

[0120] The user computing device 702 can also include one or more user input components 722 that receives user input. For example, the user input component 722 can be a touch-sensitive component (e.g., a touch-sensitive display screen or a touch pad) that is sensitive to the touch of a user input object (e.g., a finger or a stylus). The touch-sensitive component can serve to implement a virtual keyboard. Other example user input componentsinclude a microphone, a traditional keyboard, or other means by which a user can provide user input.

[0121] The server computing system 730 includes one or more processors 732 and a memory 734. The one or more processors 732 can be any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, an FPGA, a controller, a microcontroller, etc.) and can be one processor or a plurality of processors that are operatively connected. The memory 734 can include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, flash memory devices, magnetic disks, etc., and combinations thereof. The memory 734 can store data 736 and instructions 738 which are executed by the processor 732 to cause the server computing system 730 to perform operations.

[0122] In some implementations, the server computing system 730 includes or is otherwise implemented by one or more server computing devices. In instances in which the server computing system 730 includes plural server computing devices, such server computing devices can operate according to sequential computing architectures, parallel computing architectures, or some combination thereof.

[0123] As described above, the server computing system 730 can store or otherwise include one or more machine-learned models 740 (e.g., machine-learned generation models 108, machine-learned evaluation models 112, etc.). For example, the models 740 can be or can otherwise include various machine-learned models. Example machine-learned models include neural networks or other multi-layer non-linear models. Example neural networks include feed forward neural networks, deep neural networks, recurrent neural networks, and convolutional neural networks. Some example machine-learned models can leverage an attention mechanism such as self-attention. For example, some example machine-learned models can include multiheaded self-attention models (e.g., transformer models). Example models 740 are discussed with reference to Figures 1-3.

[0124] The user computing device 702 and / or the server computing system 730 can train the models 720 and / or 740 via interaction with the training computing system 750 that is communicatively coupled over the network 780. The training computing system 750 can be separate from the server computing system 730 or can be a portion of the server computing system 730.

[0125] The training computing system 750 includes one or more processors 752 and a memory 754. The one or more processors 752 can be any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, an FPGA, a controller, a microcontroller, etc.) and can be one processor or a plurality of processors that are operatively connected. The memory754 can include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, flash memory devices, magnetic disks, etc., and combinations thereof. The memory 754 can store data 756 and instructions 758 which are executed by the processor 752 to cause the training computing system 750 to perform operations. In some implementations, the training computing system 750 includes or is otherwise implemented by one or more server computing devices.

[0126] The training computing system 750 can include a model trainer 760 that trains the machine-learned models 720 and / or 740 stored at the user computing device 702 and / or the server computing system 730 using various training or learning techniques, such as, for example, backwards propagation of errors. For example, a loss function can be back propagated through the model(s) to update one or more parameters of the model(s) (e.g., based on a gradient of the loss function). Various loss functions can be used such as mean squared error, likelihood loss, cross entropy loss, hinge loss, and / or various other loss functions. Gradient descent techniques can be used to iteratively update the parameters over a number of training iterations. The training can implement supervised learning, unsupervised learning, reinforcement learning, and other training techniques.

[0127] In some implementations, performing backwards propagation of errors can include performing truncated backpropagation through time. The model trainer 760 can perform a number of generalization techniques (e.g., weight decays, dropouts, etc.) to improve the generalization capability of the models being trained.

[0128] In some implementations, the model(s) 720 can be pre-trained before domainspecific alignment. For instance, a model 720 can be pre trained over a general corpus of training data and fine-tuned on a more targeted corpus of training data. A model 720 can be aligned using prompts that are designed to elicit domain-specific outputs. Prompts can be designed to include learned prompt values (e.g., soft prompts). The trained model(s) 720 may be validated prior to their use using input data other than the training data and may be further updated or refined during their use based on additional feedback / inputs.

[0129] In particular, the model trainer 760 can train the machine-learned models 720 and / or 740 based on a set of training data 762. Training data 762 for the machine-learned generation model can include, for example, input-output pairs comprising structured insight data 106 as inputs, and example outputs 110, 118 as outputs. Training data 762 for the machine- learned evaluation model 112 can include, for example, input-output pairs comprising candidate outputs 110 as inputs, and evaluations 114 (e.g., numerical evaluation scores, etc.) as outputs. In some instances, training data 762 for the machine-learned evaluation model canalso include structured insight data 106 as part of the inputs of each input-output pair. In some instances, training data 762 for the machine-learned generation model 108 or machine-learned evaluation model 112 can include, for example, input-output pairs comprising structured insight data 106 as inputs, and feedback inputs 324 as outputs.

[0130] In some implementations, if the user has provided consent, the training examples can be provided by the user computing device 702. Thus, in such implementations, the model 720 provided to the user computing device 702 can be trained by the training computing system 750 on user-specific data received from the user computing device 702. In some instances, this process can be referred to as personalizing the model.

[0131] The model trainer 760 includes computer logic utilized to provide desired functionality. The model trainer 760 can be implemented in hardware, firmware, and / or software controlling a general purpose processor. For example, in some implementations, the model trainer 760 includes program files stored on a storage device, loaded into a memory and executed by one or more processors. In other implementations, the model trainer 760 includes one or more sets of computer-executable instructions that are stored in a tangible computer- readable storage medium such as RAM, hard disk, or optical or magnetic media.

[0132] The network 780 can be any type of communications network, such as a local area network (e.g., intranet), wide area network (e.g., Internet), or some combination thereof and can include any number of wired or wireless links. In general, communication over the network 780 can be carried via any type of wired and / or wireless connection, using a wide variety of communication protocols (e.g., TCP / IP, HTTP, SMTP, FTP), encodings or formats (e.g., HTML, XML), and / or protection schemes (e.g., VPN, secure HTTP, SSL).

[0133] Figure 7 A illustrates one example computing system that can be used to implement the present disclosure. Other computing systems can be used as well. For example, in some implementations, the user computing device 702 can include the model trainer 760 and the training dataset 762. In such implementations, the models 720 can be both trained and used locally at the user computing device 702. In some of such implementations, the user computing device 702 can implement the model trainer 760 to personalize the models 720 based on userspecific data.

[0134] Figure 7B depicts a block diagram of an example computing device 10 that performs according to example embodiments of the present disclosure. The computing device 10 can be a user computing device or a server computing device.

[0135] The computing device 10 includes a number of applications (e.g., applications 1 through N). Each application contains its own machine learning library and machine-learnedmodel(s). For example, each application can include a machine-learned model. Example applications include a text messaging application, an email application, a dictation application, a virtual keyboard application, a browser application, and other applications.

[0136] As illustrated in Figure 7B, each application can communicate with a number of other components of the computing device, such as, for example, one or more sensors, a context manager, a device state component, and / or additional components. In some implementations, each application can communicate with each device component using an API (e.g., a public API). In some implementations, the API used by each application is specific to that application.

[0137] Figure 7C depicts a block diagram of an example computing device 50 that performs according to example embodiments of the present disclosure. The computing device 50 can be a user computing device or a server computing device.

[0138] The computing device 50 includes a number of applications (e.g., applications 1 through N). Each application is in communication with a central intelligence layer. Example applications include a text messaging application, an email application, a dictation application, a virtual keyboard application, a browser application, etc. In some implementations, each application can communicate with the central intelligence layer (and model(s) stored therein) using an API (e.g., a common API across all applications).

[0139] The central intelligence layer includes a number of machine-learned models. For example, as illustrated in Figure 7C, a respective machine-learned model can be provided for each application and managed by the central intelligence layer. In other implementations, two or more applications can share a single machine-learned model. For example, in some implementations, the central intelligence layer can provide a single model for all of the applications. In some implementations, the central intelligence layer is included within or otherwise implemented by an operating system of the computing device 50.

[0140] The central intelligence layer can communicate with a central device data layer. The central device data layer can be a centralized repository of data for the computing device 50. As illustrated in Figure 7C, the central device data layer can communicate with a number of other components of the computing device, such as, for example, one or more sensors, a context manager, a device state component, and / or additional components. In some implementations, the central device data layer can communicate with each device component using an API (e.g., a private API).

[0141] Example Training Method

[0142] Figure 8 depicts a flowchart of a method 800 for training one or more machine- learned models according to aspects of the present disclosure. For instance, an example machine-learned model can include a machine-learned model 235.

[0143] One or more portion(s) of example method 800 can be implemented by a computing system that includes one or more computing devices such as, for example, computing systems described with reference to the other figures. Each respective portion of example method 800 can be performed by any (or any combination) of one or more computing devices. Moreover, one or more portion(s) of example method 800 can be implemented on the hardware components of the device(s) described herein, for example, to train one or more systems or models. Figure 8 depicts elements performed in a particular order for purposes of illustration and discussion. Those of ordinary skill in the art, using the disclosures provided herein, will understand that the elements of any of the methods discussed herein can be adapted, rearranged, expanded, omitted, combined, or modified in various ways without deviating from the scope of the present disclosure. Figure 8 is described with reference to elements / terms described with respect to other systems and figures for exemplary illustrated purposes and is not meant to be limiting. One or more portions of example method 800 can be performed additionally, or alternatively, by other systems.

[0144] At 802, example method 800 can include obtaining a training instance. A set of training data can include a plurality of training instances divided between multiple datasets (e.g., a training dataset, a validation dataset, or testing dataset). A training instance can be labeled or unlabeled. Although referred to in example method 800 as a “training” instance, it is to be understood that runtime inferences can form training instances when a model is trained using an evaluation of the model’s performance on that runtime instance (e.g., online training / learning). Example data types for the training instance and various tasks associated therewith are described throughout the present disclosure.

[0145] At 804, example method 800 can include processing, using one or more machine- learned models, the training instance to generate an output. The output can be directly obtained from the one or more machine-learned models or can be a downstream result of a chain of processing operations that includes an output of the one or more machine-learned models.

[0146] At 806, example method 800 can include receiving an evaluation signal associated with the output. The evaluation signal can be obtained using a loss function. Various determinations of loss can be used, such as mean squared error, likelihood loss, cross entropy loss, hinge loss, contrastive loss, or various other loss functions. The evaluation signal can be computed using known ground-truth labels (e.g., supervised learning), predicted or estimatedlabels (e.g., semi- or self-supervised learning), or without labels (e.g., unsupervised learning). The evaluation signal can be a reward (e.g., for reinforcement learning). The reward can be computed using a machine-learned reward model configured to generate rewards based on output(s) received. The reward can be computed using feedback data describing human feedback on the output(s).

[0147] At 808, example method 800 can include updating the machine-learned model using the evaluation signal. For example, values for parameters of the machine-learned model(s) can be learned, in some embodiments, using various training or learning techniques, such as, for example, backwards propagation. For example, the evaluation signal can be back propagated from the output (or another source of the evaluation signal) through the machine- learned model(s) to update one or more parameters of the model(s) (e.g., based on a gradient of the evaluation signal with respect to the parameter value(s)). For example, system(s) containing one or more machine-learned models can be trained in an end-to-end manner. Gradient descent techniques can be used to iteratively update the parameters over a number of training iterations. In some implementations, performing backwards propagation of errors can include performing truncated b ackpropagation through time. Example method 800 can include implementing a number of generalization techniques (e.g., weight decays, dropouts, etc.) to improve the generalization capability of the models being trained.

[0148] In some implementations, example method 800 can be implemented for training a machine-learned model from an initialized state to a fully trained state (e.g., when the model exhibits a desired performance profile, such as based on accuracy, precision, recall, etc.).

[0149] In some implementations, example method 800 can be implemented for particular stages of a training procedure. For instance, in some implementations, example method 800 can be implemented for pre-training a machine-learned model. Pre-training can include, for instance, large-scale training over potentially noisy data to achieve a broad base of performance levels across a variety of tasks / data types. In some implementations, example method 800 can be implemented for fine-tuning a machine-learned model. Fine-tuning can include, for instance, smaller-scale training on higher-quality (e.g., labeled, curated, etc.) data. Fine-tuning can affect all or a portion of the parameters of a machine-learned model. For example, various portions of the machine-learned model can be “frozen” for certain training stages. For example, parameters associated with an embedding space can be “frozen” during fine-tuning (e.g., to retain information learned from a broader domain(s) than present in the fine-tuning dataset(s)). An example fine-tuning approach includes reinforcement learning. Reinforcement learning can be based on user feedback on model performance during use.Example Machine-Learned Models

[0150] Figure 9 is a block diagram of an example processing flow for using machine- learned model(s) 1 to process input(s) 2 to generate output(s) 3.

[0151] Machine-learned model(s) 1 can be or include one or multiple machine-learned models or model components. Example machine-learned models can include neural networks (e.g., deep neural networks). Example machine-learned models can include non-linear models or linear models. Example machine-learned models can use other architectures in lieu of or in addition to neural networks. Example machine-learned models can include decision tree based models, support vector machines, hidden Markov models, Bayesian networks, linear regression models, k-means clustering models, etc.

[0152] Example neural networks can include feed-forward neural networks, recurrent neural networks (RNNs), including long short-term memory (LSTM) based recurrent neural networks, convolutional neural networks (CNNs), diffusion models, generative-adversarial networks, or other forms of neural networks. Example neural networks can be deep neural networks. Some example machine-learned models can leverage an attention mechanism such as self-attention. For example, some example machine-learned models can include multiheaded self-attention models.

[0153] Machine-learned model(s) 1 can include a single or multiple instances of the same model configured to operate on data from input(s) 2. Machine-learned model(s) 1 can include an ensemble of different models that can cooperatively interact to process data from input(s) 2. For example, machine-learned model(s) 1 can employ a mixture-of-experts structure.

[0154] Input(s) 2 can generally include or otherwise represent various types of data. Input(s) 2 can include one type or many different types of data. Output(s) 3 can be data of the same type(s) or of different types of data as compared to input(s) 2. Output(s) 3 can include one type or many different types of data.

[0155] Example data types for input(s) 2 or output(s) 3 include natural language text data, software code data (e.g., source code, object code, machine code, or any other form of computer-readable instructions or programming languages), machine code data (e.g., binary code, assembly code, or other forms of machine-readable instructions that can be executed directly by a computer's central processing unit), assembly code data (e.g., low-level programming languages that use symbolic representations of machine code instructions to program a processing unit), genetic data or other chemical or biochemical data, image data, audio data, audiovisual data, haptic data, biometric data, medical data, financial data, statistical data, geographical data, astronomical data, historical data, sensor data generally (e.g., digitalor analog values, such as voltage or other absolute or relative level measurement values from a real or artificial input, such as from an audio sensor, light sensor, displacement sensor, etc.), and the like. Data can be raw or processed and can be in any format or schema.

[0156] In multimodal inputs 2 or outputs 3, example combinations of data types include image data and audio data, image data and natural language data, natural language data and software code data, image data and biometric data, sensor data and medical data, etc. It is to be understood that any combination of data types in an input 2 or an output 3 can be present.

[0157] An example input 2 can include one or multiple data types, such as the example data types noted above. An example output 3 can include one or multiple data types, such as the example data types noted above. The data type(s) of input 2 can be the same as or different from the data type(s) of output 3. It is to be understood that the example data types noted above are provided for illustrative purposes only. Data types contemplated within the scope of the present disclosure are not limited to those examples noted above.

[0158] Example Machine-Learned Sequence Processing Models

[0159] Figure 10 is a block diagram of an example implementation of an example machine-learned model configured to process sequences of information. For instance, an example implementation of machine-learned model(s) 1 can include machine-learned sequence processing model(s) 4. An example system can pass input(s) 2 to sequence processing model(s) 4. Sequence processing model(s) 4 can include one or more machine-learned components. Sequence processing model(s) 4 can process the data from input(s) 2 to obtain an input sequence 5. Input sequence 5 can include one or more input elements 5-1, 5-2, . . . , 5-M, etc. obtained from input(s) 2. Sequence processing model 4 can process input sequence 5 using prediction layer(s) 6 to generate an output sequence 7. Output sequence 7 can include one or more output elements 7-1, 7-2, . . . , 7-N, etc. generated based on input sequence 5. The system can generate output(s) 3 based on output sequence 7.

[0160] Sequence processing model(s) 4 can include one or multiple machine-learned model components configured to ingest, generate, or otherwise reason over sequences of information. For example, some example sequence processing models in the text domain are referred to as Large Language Models (LLMs). Other example sequence processing models can operate in other domains, such as image domains, audio domains, biochemical domains, by way of example. Sequence processing model(s) 4 can process one or multiple types of data simultaneously. Sequence processing model(s) 4 can include relatively large models (e.g., moreparameters, computationally expensive, etc.), relatively small models (e.g., fewer parameters, computationally lightweight, etc.), or both.

[0161] In general, sequence processing model(s) 4 can obtain input sequence 5 using data from input(s) 2. For instance, input sequence 5 can include a representation of data from input(s) 2 in a format understood by sequence processing model(s) 4. One or more machine- learned components of sequence processing model(s) 4 can ingest the data from input(s) 2, parse the data into pieces compatible with the processing architectures of sequence processing model(s) 4 (e.g., via tokenization), and project the pieces into an input space associated with prediction layer(s) 6 (e.g., via embedding).

[0162] Sequence processing model(s) 4 can ingest the data from input(s) 2 and parse the data into a sequence of elements to obtain input sequence 5. For example, a portion of input data from input(s) 2 can be broken down into pieces that collectively represent the content of the portion of the input data. The pieces can provide the elements of the sequence.

[0163] Elements 5-1, 5-2, . . . , 5-M can represent, in some cases, building blocks for capturing or expressing meaningful information in a particular data domain. For instance, the elements can describe “atomic units” across one or more domains. For example, for textual input source(s), the elements can correspond to groups of one or more words or sub-word components, such as sets of one or more characters.

[0164] For example, elements 5-1, 5-2, . . . , 5-M can represent tokens obtained using a tokenizer. For instance, a tokenizer can process a given portion of an input source and output a series of tokens (e.g., corresponding to input elements 5-1, 5-2, . . . , 5-M) that represent the portion of the input source. Various approaches to tokenization can be used. For instance, textual input source(s) can be tokenized using a byte-pair encoding (BPE) technique. Imagebased input source(s) can be tokenized by extracting and serializing patches from an image.

[0165] In general, arbitrary data types can be serialized and processed into input sequence 5. It is to be understood that element(s) 5-1, 5-2, . . . , 5-M depicted in Figure 10 can be the tokens or can be the embedded representations thereof.

[0166] Prediction layer(s) 6 can predict one or more output elements 7-1, 7-2, . . . , 7-N based on the input elements. Prediction layer(s) 6 can include one or more machine-learned model architectures, such as one or more layers of learned parameters that manipulate and transform the input(s) to extract higher-order meaning from, and relationships between, input element(s) 5-1, 5-2, . . . , 5-M. In this manner, for instance, example prediction layer(s) 6 can predict new output element(s) in view of the context provided by input sequence 5.

[0167] Prediction layer(s) 6 can evaluate associations between portions of input sequence5 and a particular output element. These associations can inform a prediction of the likelihood that a particular output follows the input context. For example, consider the textual snippet, “The carpenter’s toolbox was small and heavy. It was full of .” Example prediction layer(s)6 can identify that “It” refers back to “toolbox” by determining a relationship between the respective embeddings. Example prediction layer(s) 6 can also link “It” to the attributes of the toolbox, such as “small” and “heavy.” Based on these associations, prediction layer(s) 6 can, for instance, assign a higher probability to the word “nails” than to the word “sawdust.”

[0168] A transformer is an example architecture that can be used in prediction layer(s) 4. A transformer is an example of a machine-learned model architecture that uses an attention mechanism to compute associations between items within a context window. The context window can include a sequence that contains input sequence 5 and potentially one or more output element(s) 7-1, 7-2, . . . , 7-N. A transformer block can include one or more attention layer(s) and one or more post-attention layer(s) (e.g., feedforward layer(s), such as a multilayer perceptron).

[0169] Prediction layer(s) 6 can include other machine-learned model architectures in addition to or in lieu of transformer-based architectures. For example, recurrent neural networks (RNNs) and long short-term memory (LSTM) models can also be used, as well as convolutional neural networks (CNNs). In general, prediction layer(s) 6 can leverage various kinds of artificial neural networks that can understand or generate sequences of information.

[0170] Output sequence 7 can include or otherwise represent the same or different data types as input sequence 5. For instance, input sequence 5 can represent textual data, and output sequence 7 can represent textual data. Input sequence 5 can represent image, audio, or audiovisual data, and output sequence 7 can represent textual data (e.g., describing the image, audio, or audiovisual data). It is to be understood that prediction layer(s) 6, and any other interstitial model components of sequence processing model(s) 4, can be configured to receive a variety of data types in input sequence(s) 5 and output a variety of data types in output sequence(s) 7.

[0171] Output sequence 7 can have various relationships to input sequence 5. Output sequence 7 can be a continuation of input sequence 5. Output sequence 7 can be complementary to input sequence 5. Output sequence 7 can translate, transform, augment, or otherwise modify input sequence 5. Output sequence 7 can answer, evaluate, confirm, or otherwise respond to input sequence 5. Output sequence 7 can implement (or describe instructions for implementing) an instruction provided via input sequence 5.

[0172] Output sequence 7 can be generated autoregressive. For instance, for some applications, an output of one or more prediction layer(s) 6 can be passed through one or more output layers (e.g., softmax layer) to obtain a probability distribution over an output vocabulary (e.g., a textual or symbolic vocabulary) conditioned on a set of input elements in a context window. In this manner, for instance, output sequence 7 can be autoregressive generated by sampling a likely next output element, adding that element to the context window, and regenerating the probability distribution based on the updated context window, and sampling a likely next output element, and so forth.

[0173] Output sequence 7 can also be generated non-autoregressive. For instance, multiple output elements of output sequence 7 can be predicted together without explicit sequential conditioning on each other. Output sequence 7 can include one or multiple portions or elements. In an example content generation configuration, output sequence 7 can include multiple elements corresponding to multiple portions of a generated output sequence (e.g., a textual sentence, values of a discretized waveform, computer code, etc.). In an example classification configuration, output sequence 7 can include a single element associated with a classification output. For instance, an output “vocabulary” can include a set of classes into which an input sequence is to be classified. For instance, a vision transformer block can pass latent state information to a multilayer perceptron that outputs a likely class value associated with an input image.

[0174] Figure 11 is a block diagram of an example technique for populating an example input sequence 8. Input sequence 8 can include various functional elements that form part of the model infrastructure, such as an element 8-0 obtained from a task indicator 9 that signals to any model(s) that process input sequence 8 that a particular task is being performed (e.g., to help adapt a performance of the model(s) to that particular task). Input sequence 8 can include various data elements from different data modalities. For instance, an input modality 10-1 can include one modality of data. A data-to-sequence model 11-1 can process data from input modality 10-1 to project the data into a format compatible with input sequence 8 (e.g., one or more vectors dimensioned according to the dimensions of input sequence 8) to obtain elements 8-1, 8-2, 8-3. Another input modality 10-2 can include a different modality of data. A data-to- sequence model 11-2 can project data from input modality 10-2 into a format compatible with input sequence 8 to obtain elements 8-4, 8-5, 8-6. Another input modality 10-3 can include yet another different modality of data. A data-to-sequence model 11-3 can project data from input modality 10-3 into a format compatible with input sequence 8 to obtain elements 8-7, 8-8, 8-9.

[0175] Input sequence 8 can be the same as or different from input sequence 5. Input sequence 8 can be a multimodal input sequence that contains elements that represent data from different modalities using a common dimensional representation. For instance, an embedding space can have P dimensions. Input sequence 8 can be configured to contain a plurality of elements that have P dimensions. In this manner, for instance, example implementations can facilitate information extraction and reasoning across diverse data modalities by projecting data into elements in the same embedding space for comparison, combination, or other computations therebetween.

[0176] For example, elements 8-0, . . . , 8-9 can indicate particular locations within a multidimensional embedding space. Some elements can map to a set of discrete locations in the embedding space. For instance, elements that correspond to discrete members of a predetermined vocabulary of tokens can map to discrete locations in the embedding space that are associated with those tokens. Other elements can be continuously distributed across the embedding space. For instance, some data types can be broken down into continuously defined portions (e.g., image patches) that can be described using continuously distributed locations within the embedding space.

[0177] In some implementations, the expressive power of the embedding space may not be limited to meanings associated with any particular set of tokens or other building blocks. For example, a continuous embedding space can encode a spectrum of high-order information. An individual piece of information (e.g., a token) can map to a particular point in that space: for instance, a token for the word “dog” can be projected to an embedded value that points to a particular location in the embedding space associated with canine-related information. Similarly, an image patch of an image of a dog on grass can also be projected into the embedding space. In some implementations, the projection of the image of the dog can be similar to the projection of the word “dog” while also having similarity to a projection of the word “grass,” while potentially being different from both. In some implementations, the projection of the image patch may not exactly align with any single projection of a single word. In some implementations, the projection of the image patch can align with a combination of the projections of the words “dog” and “grass.” In this manner, for instance, a high-order embedding space can encode information that can be independent of data modalities in which the information is expressed.

[0178] Task indicator 9 can include a model or model component configured to identify a task being performed and inject, into input sequence 8, an input value represented by element 8-0 that signals which task is being performed. For instance, the input value can be providedas a data type associated with an input modality and projected along with that input modality (e.g., the input value can be a textual task label that is embedded along with other textual data in the input; the input value can be a pixel-based representation of a task that is embedded along with other image data in the input; etc.). The input value can be provided as a data type that differs from or is at least independent from other input(s). For instance, the input value represented by element 8-0 can be learned within a continuous embedding space.

[0179] Input modalities 10-1, 10-2, and 10-3 can be associated with various different data types (e.g., as described above with respect to input(s) 2 and output(s) 3).

[0180] Data-to-sequence models 11-1, 11-2, and 11-3 can be the same or different from each other. Data-to-sequence models 11-1, 11-2, and 11-3 can be adapted to each respective input modality 10-1, 10-2, and 10-3. For example, a textual data-to-sequence model can subdivide a portion of input text and project the subdivisions into element(s) in input sequence 8 (e.g., elements 8-1, 8-2, 8-3, etc.). An image data-to-sequence model can subdivide an input image and project the subdivisions into element(s) in input sequence 8 (e.g., elements 8-4, 8- 5, 8-6, etc.). An arbitrary data type data-to-sequence model can subdivide an input of that arbitrary data type and project the subdivisions into element(s) in input sequence 8 (e.g., elements 8-7, 8-8, 8-9, etc.).

[0181] Data-to-sequence models 11-1, 11-2, and 11-3 can form part of machine-learned sequence processing model(s) 4. Data-to-sequence models 11-1, 11-2, and 11-3 can be jointly trained with or trained independently from machine-learned sequence processing model(s) 4. Data-to-sequence models 11-1, 11-2, and 11-3 can be trained end-to-end with machine-learned sequence processing model(s) 4.Example Machine-Learned Model Development Platform

[0182] Figure 12 is a block diagram of an example model development platform 12 that can facilitate creation, adaptation, and refinement of example machine-learned models (e.g., machine-learned model(s) 1, sequence processing model(s) 4, etc.). Model development platform 12 can provide a number of different toolkits that developer systems can employ in the development of new or adapted machine-learned models.

[0183] Model development platform 12 can provide one or more model libraries 13 containing building blocks for new models. Model libraries 13 can include one or more pretrained foundational models 13-1, which can provide a backbone of processing power across various tasks. Model libraries 13 can include one or more pre-trained expert models 13-2, which can be focused on performance in particular domains of expertise. Model libraries 13can include various model primitives 13-3, which can provide low-level architectures or components (optionally pre-trained), which can be assembled in various arrangements as desired.

[0184] Model development platform 12 can receive selections of various model components 14. Model development platform 12 can pass selected model components 14 to a workbench 15 that combines selected model components 14 into a development model 16.

[0185] Workbench 15 can facilitate further refinement and adaptation of development model 16 by leveraging a number of different toolkits integrated with model development platform 12. For example, workbench 15 can facilitate alignment of the development model 16 with a desired performance profile on various tasks using a model alignment toolkit 17.

[0186] Model alignment toolkit 17 can provide a number of tools for causing development model 16 to generate outputs aligned with desired behavioral characteristics. Alignment can include increasing accuracy, precision, and / or recall of model outputs. Alignment can include enforcing output styles, schema, or other preferential characteristics of model outputs. Alignment can be general or domain-specific. For instance, a pre-trained foundational model 13-1 can begin with an initial level of performance across multiple domains. Alignment of the pre-trained foundational model 13-1 can include improving a performance in a particular domain of information or tasks (e.g., even at the expense of performance in another domain of information or tasks).

[0187] Model alignment toolkit 17 can integrate one or more dataset(s) 17-1 for aligning development model 16. Curated dataset(s) 17-1 can include labeled or unlabeled training data. Dataset(s) 17-1 can be obtained from public domain datasets. Dataset(s) 17-1 can be obtained from private datasets associated with one or more developer system(s) for the alignment of bespoke machine-learned model(s) customized for private use-cases.

[0188] Pre-training pipelines 17-2 can include a machine-learned model training workflow configured to update development model 16 over large-scale, potentially noisy datasets. For example, pre-training can leverage unsupervised learning techniques (e.g., denoising, etc.) to process large numbers of training instances to update model parameters from an initialized state and achieve a desired baseline performance. Pre-training pipelines 17-2 can leverage unlabeled datasets in dataset(s) 17-1 to perform pre-training. Workbench 15 can implement a pre-training pipeline 17-2 to pre-train development model 16.

[0189] Fine-tuning pipelines 17-3 can include a machine-learned model training workflow configured to refine the model parameters of development model 16 with higher-quality data. Fine-tuning pipelines 17-3 can update development model 16 by conducting supervisedtraining with labeled dataset(s) in dataset(s) 17-1. Fine-tuning pipelines 17-3 can update development model 16 by conducting reinforcement learning using reward signals from user feedback signals. Workbench 15 can implement a fine-tuning pipeline 17-3 to fine-tune development model 16.

[0190] Prompt libraries 17-4 can include sets of inputs configured to induce behavior aligned with desired performance criteria. Prompt libraries 17-4 can include few-shot prompts (e.g., inputs providing examples of desired model outputs for prepending to a desired runtime query), chain-of-thought prompts (e.g., inputs providing step-by-step reasoning within the exemplars to facilitate thorough reasoning by the model), and the like.

[0191] Example prompts can be retrieved from an available repository of prompt libraries 17-4. Example prompts can be contributed by one or more developer systems using workbench 15.

[0192] In some implementations, pre-trained or fine-tuned models can achieve satisfactory performance without exemplars in the inputs. For instance, zero-shot prompts can include inputs that lack exemplars. Zero-shot prompts can be within a domain within a training dataset or outside of the training domain(s).

[0193] Prompt libraries 17-4 can include one or more prompt engineering tools. Prompt engineering tools can provide workflows for retrieving or learning optimized prompt values. Prompt engineering tools can facilitate directly learning prompt values (e.g., input element values) based one or more training iterations. Workbench 15 can implement prompt engineering tools in development model 16.

[0194] Prompt libraries 17-4 can include pipelines for prompt generation. For example, inputs can be generated using development model 16 itself or other machine-learned models. In this manner, for instance, a first model can process information about a task and output an input for a second model to process in order to perform a step of the task. The second model can be the same as or different from the first model. Workbench 15 can implement prompt generation pipelines in development model 16.

[0195] Prompt libraries 17-4 can include pipelines for context injection. For instance, a performance of development model 16 on a particular task can improve if provided with additional context for performing the task. Prompt libraries 17-4 can include software components configured to identify desired context, retrieve the context from an external source (e.g., a database, a sensor, etc.), and add the context to the input prompt. Workbench 15 can implement context injection pipelines in development model 16.

[0196] Although various training examples described herein with respect to model development platform 12 refer to “pre-training” and “fine-tuning,” it is to be understood that model alignment toolkit 17 can generally support a wide variety of training techniques adapted for training a wide variety of machine-learned models. Example training techniques can correspond to the example training method 800 described above.

[0197] Model development platform 12 can include a model plugin toolkit 18. Model plugin toolkit 18 can include a variety of tools configured for augmenting the functionality of a machine-learned model by integrating the machine-learned model with other systems, devices, and software components. For instance, a machine-learned model can use tools to increase performance quality where appropriate. For instance, deterministic tasks can be offloaded to dedicated tools in lieu of probabilistically performing the task with an increased risk of error. For instance, instead of autoregressively predicting the solution to a system of equations, a machine-learned model can recognize a tool to call for obtaining the solution and pass the system of equations to the appropriate tool. The tool can be a traditional system of equations solver that can operate deterministically to resolve the system of equations. The output of the tool can be returned in response to the original query. In this manner, tool use can allow some example models to focus on the strengths of machine-learned models — e.g., understanding an intent in an unstructured request for a task — while augmenting the performance of the model by offloading certain tasks to a more focused tool for rote application of deterministic algorithms to a well-defined problem.

[0198] Model plugin toolkit 18 can include validation tools 18-1. Validation tools 18-1 can include tools that can parse and confirm output(s) of a machine-learned model. Validation tools 18-1 can include engineered heuristics that establish certain thresholds applied to model outputs. For example, validation tools 18-1 can ground the outputs of machine-learned models to structured data sources (e.g., to mitigate “hallucinations”).

[0199] Model plugin toolkit 18 can include tooling packages 18-2 for implementing one or more tools that can include scripts or other executable code that can be executed alongside development model 16. Tooling packages 18-2 can include one or more inputs configured to cause machine-learned model(s) to implement the tools (e.g., few-shot prompts that induce a model to output tool calls in the proper syntax, etc.). Tooling packages 18-2 can include, for instance, fine-tuning training data for training a model to use a tool.

[0200] Model plugin toolkit 18 can include interfaces for calling external application programming interfaces (APIs) 18-3. For instance, in addition to or in lieu of implementingtool calls or tool code directly with development model 16, development model 16 can be aligned to output instructions that initiate API calls to send or obtain data via external systems.

[0201] Model plugin toolkit 18 can integrate with prompt libraries 17-4 to build a catalog of available tools for use with development model 16. For instance, a model can receive, in an input, a catalog of available tools, and the model can generate an output that selects a tool from the available tools and initiates a tool call for using the tool.

[0202] Model development platform 12 can include a computational optimization toolkit 19 for optimizing a computational performance of development model 16. For instance, tools for model compression 19-1 can allow development model 16 to be reduced in size while maintaining a desired level of performance. For instance, model compression 19-1 can include quantization workflows, weight pruning and sparsification techniques, etc. Tools for hardware acceleration 19-2 can facilitate the configuration of the model storage and execution formats to operate optimally on different hardware resources. For instance, hardware acceleration 19- 2 can include tools for optimally sharding models for distributed processing over multiple processing units for increased bandwidth, lower unified memory requirements, etc. Tools for distillation 19-3 can provide for the training of lighter-weight models based on the knowledge encoded in development model 16. For instance, development model 16 can be a highly performant, large machine-learned model optimized using model development platform 12. To obtain a lightweight model for running in resource-constrained environments, a smaller model can be a “student model” that learns to imitate development model 16 as a “teacher model.” In this manner, for instance, the investment in learning the parameters and configurations of development model 16 can be efficiently transferred to a smaller model for more efficient inference.

[0203] Workbench 15 can implement one, multiple, or none of the toolkits implemented in model development platform 12. Workbench 15 can output an output model 20 based on development model 16. Output model 20 can be a deployment version of development model 16. Output model 20 can be a development or training checkpoint of development model 16. Output model 20 can be a distilled, compressed, or otherwise optimized version of development model 16.

[0204] Figure 13 is a block diagram of an example training flow for training a machine- learned development model 16. One or more portion(s) of the example training flow can be implemented by a computing system that includes one or more computing devices such as, for example, computing systems described with reference to the other figures. Each respective portion of the example training flow can be performed by any (or any combination) of one ormore computing devices. Moreover, one or more portion(s) of the example training flow can be implemented on the hardware components of the device(s) described herein, for example, to train one or more systems or models. FIG. 13 depicts elements performed in a particular order for purposes of illustration and discussion. Those of ordinary skill in the art, using the disclosures provided herein, will understand that the elements of any of the methods discussed herein can be adapted, rearranged, expanded, omitted, combined, or modified in various ways without deviating from the scope of the present disclosure. FIG. 13 is described with reference to elements / terms described with respect to other systems and figures for exemplary illustrated purposes and is not meant to be limiting. One or more portions of the example training flow can be performed additionally, or alternatively, by other systems.

[0205] Initially, development model 16 can persist in an initial state as an initialized model 21. Development model 16 can be initialized with weight values. Initial weight values can be random or based on an initialization schema. Initial weight values can be based on prior pretraining for the same or for a different model.

[0206] Initialized model 21 can undergo pre-training in a pre-training stage 22. Pretraining stage 22 can be implemented using one or more pre-training pipelines 17-2 over data from dataset(s) 17-1. Pre-training can be omitted, for example, if initialized model 21 is already pre-trained (e.g., development model 16 contains, is, or is based on a pre-trained foundational model or an expert model).

[0207] Pre-trained model 23 can then be a new version of development model 16, which can persist as development model 16 or as a new development model. Pre-trained model 23 can be the initial state if development model 16 was already pre-trained. Pre-trained model 23 can undergo fine-tuning in a fine-tuning stage 24. Fine-tuning stage 24 can be implemented using one or more fine-tuning pipelines 17-3 over data from dataset(s) 17-1. Fine-tuning can be omitted, for example, if a pre-trained model has satisfactory performance, if the model was already fine-tuned, or if other tuning approaches are preferred.

[0208] Fine-tuned model 25 can then be a new version of development model 16, which can persist as development model 16 or as a new development model. Fine-tuned model 25 can be the initial state if development model 16 was already fine-tuned. Fine-tuned model 25 can undergo refinement with user feedback 26. For instance, refinement with user feedback 26 can include reinforcement learning, optionally based on human feedback from human users of finetuned model 25. As reinforcement learning can be a form of fine-tuning, it is to be understood that fine-tuning stage 24 can subsume the stage for refining with user feedback 26. Refinementwith user feedback 26 can produce a refined model 27. Refined model 27 can be output to downstream system(s) 28 for deployment or further development.

[0209] In some implementations, computational optimization operations can be applied before, during, or after each stage. For instance, initialized model 21 can undergo computational optimization 29-1 (e.g., using computational optimization toolkit 19) before pretraining stage 22. Pre-trained model 23 can undergo computational optimization 29-2 (e.g., using computational optimization toolkit 19) before fine-tuning stage 24. Fine-tuned model 25 can undergo computational optimization 29-3 (e.g., using computational optimization toolkit 19) before refinement with user feedback 26. Refined model 27 can undergo computational optimization 29-4 (e.g., using computational optimization toolkit 19) before output to downstream system(s) 28. Computational optimization(s) 29-1, . . . , 29-4 can all be the same, all be different, or include at least some different optimization techniques.Example Machine-Learned Model Inference System

[0210] Figure 14 is a block diagram of an inference system for operating one or more machine-learned model(s) 1 to perform inference (e.g., for training, for deployment, etc.). A model host 31 can receive machine-learned model(s) 1. Model host 31 can host one or more model instance(s) 31-1, which can be one or multiple instances of one or multiple models. Model host 31 can host model instance(s) 31-1 using available compute resources 31-2 associated with model host 31.

[0211] Model host 31 can perform inference on behalf of one or more client(s) 32. Client(s) 32 can transmit an input request 33 to model host 31. Using input request 33, model host 31 can obtain input(s) 2 for input to machine-learned model(s) 1. Machine-learned model(s) 1 can process input(s) 2 to generate output(s) 3. Using output(s) 3, model host 31 can return an output payload 34 for responding to input request 33 from client(s) 32. Output payload 34 can include or be based on output(s) 3.

[0212] Model host 31 can leverage various other resources and tools to augment the inference task. For instance, model host 31 can communicate with tool interfaces 35 to facilitate tool use by model instance(s) 31-1. Tool interfaces 35 can include local or remote APIs. Tool interfaces 35 can include integrated scripts or other software functionality. Model host 31 can engage online learning interface(s) 36 to facilitate ongoing improvements to machine-learned model(s) 1. For instance, online learning interface(s) 36 can be used within reinforcement learning loops to retrieve user feedback on inferences served by model host 31. Model host 31 can access runtime data source(s) 37 for augmenting input(s) 2 with additional contextualinformation. For instance, runtime data source(s) 37 can include a knowledge graph 37-1 that facilitates structured information retrieval for information associated with input request(s) 33 (e.g., a search engine service). According to some embodiments, the knowledge graph 37-1 can include a plurality of structured data entries associated with a user. In some instances, the machine-learned model(s) 1 can process a structured data entry with the knowledge graph to determine an insight about the user. The structured data entry can be generated by Runtime data source(s) 37 can include public or private, external or local database(s) 37-2 that can store information associated with input request(s) 33 for augmenting input(s) 2. Runtime data source(s) 37 can include account data 37-3 which can be retrieved in association with a user account corresponding to a client 32 for customizing the behavior of model host 31 accordingly.

[0213] Model host 31 can be implemented by one or multiple computing devices or systems. Client(s) 2 can be implemented by one or multiple computing devices or systems, which can include computing devices or systems shared with model host 31.

[0214] For example, model host 31 can operate on a server system that provides a machine-learning service to client device(s) that operate client(s) 32 (e.g., over a local or wide- area network). Client device(s) can be end-user devices used by individuals. Client device(s) can be server systems that operate client(s) 32 to provide various functionality as a service to downstream end-user devices.

[0215] In some implementations, model host 31 can operate on the same device or system as client(s) 32. Model host 31 can be a machine-learning service that runs on-device to provide machine-learning functionality to one or multiple applications operating on a client device, which can include an application implementing client(s) 32. Model host 31 can be a part of the same application as client(s) 32. For instance, model host 31 can be a subroutine or method implemented by one part of an application, and client(s) 32 can be another subroutine or method that engages model host 31 to perform inference functions within the application. It is to be understood that model host 31 and client(s) 32 can have various different configurations.

[0216] Model instance(s) 31-1 can include one or more machine-learned models that are available for performing inference. Model instance(s) 31-1 can include weights or other model components that are stored in persistent storage, temporarily cached, or loaded into high-speed memory. Model instance(s) 31-1 can include multiple instance(s) of the same model (e.g., for parallel execution of more requests on the same model). Model instance(s) 31-1 can include instance(s) of different model(s). Model instance(s) 31-1 can include cached intermediate states of active or inactive model(s) used to accelerate inference of those models. For instance, aninference session with a particular model may generate significant amounts of computational results that can be re-used for future inference runs (e.g., using a KV cache for transformerbased models). These computational results can be saved in association with that inference session so that session can be executed more efficiently when resumed.

[0217] Compute resource(s) 31-2 can include one or more processors (central processing units, graphical processing units, tensor processing units, machine-learning accelerators, etc.) connected to one or more memory devices. Compute resource(s) 31-2 can include a dynamic pool of available resources shared with other processes. Compute resource(s) 31-2 can include memory devices large enough to fit an entire model instance in a single memory instance. Compute resource(s) 31-2 can also shard model instance(s) across multiple memory devices (e.g., using data parallelization or tensor parallelization, etc.). This can be done to increase parallelization or to execute a large model using multiple memory devices which individually might not be able to fit the entire model into memory.

[0218] Input request 33 can include data for input(s) 2. Model host 31 can process input request 33 to obtain input(s) 2. Input(s) 2 can be obtained directly from input request 33 or can be retrieved using input request 33. Input request 33 can be submitted to model host 31 via an API.

[0219] Model host 31 can perform inference over batches of input requests 33 in parallel. For instance, a model instance 31-1 can be configured with an input structure that has a batch dimension. Separate input(s) 2 can be distributed across the batch dimension (e.g., rows of an array). The separate input(s) 2 can include completely different contexts. The separate input(s) 2 can be multiple inference steps of the same task. The separate input(s) 2 can be staggered in an input structure, such that any given inference cycle can be operating on different portions of the respective input(s) 2. In this manner, for instance, model host 31 can perform inference on the batch in parallel, such that output(s) 3 can also contain the batch dimension and return the inference results for the batched input(s) 2 in parallel. In this manner, for instance, batches of input request(s) 33 can be processed in parallel for higher throughput of output payload(s) 34.

[0220] Output payload 34 can include or be based on output(s) 3 from machine-learned model(s) 1. Model host 31 can process output(s) 3 to obtain output payload 34. This can include chaining multiple rounds of inference (e.g., iteratively, recursively, across the same model(s) or different model(s)) to arrive at a final output for a task to be returned in output payload 34. Output payload 34 can be transmitted to client(s) 32 via an API.

[0221] Online learning interface(s) 36 can facilitate reinforcement learning of machine- learned model(s) 1. Online learning interface(s) 36 can facilitate reinforcement learning withhuman feedback (RLHF). Online learning interface(s) 36 can facilitate federated learning of machine-learned model(s) 1.

[0222] Model host 31 can execute machine-learned model(s) 1 to perform inference for various tasks using various types of data. For example, various different input(s) 2 and output(s) 3 can be used for various different tasks. In some implementations, input(s) 2 can be or otherwise represent image data. Machine-learned model(s) 1 can process the image data to generate an output. As an example, machine-learned model(s) 1 can process the image data to generate an image recognition output (e.g., a recognition of the image data, a latent embedding of the image data, an encoded representation of the image data, a hash of the image data, etc.). As another example, machine-learned model(s) 1 can process the image data to generate an image segmentation output. As another example, machine-learned model(s) 1 can process the image data to generate an image classification output. As another example, machine-learned model(s) 1 can process the image data to generate an image data modification output (e.g., an alteration of the image data, etc.). As another example, machine-learned model(s) 1 can process the image data to generate an encoded image data output (e.g., an encoded and / or compressed representation of the image data, etc.). As another example, machine-learned model(s) 1 can process the image data to generate an upscaled image data output. As another example, machine-learned model(s) 1 can process the image data to generate a prediction output.

[0223] In some implementations, the task is a computer vision task. In some cases, input(s) 2 includes pixel data for one or more images and the task is an image processing task. For example, the image processing task can be image classification, where the output is a set of scores, each score corresponding to a different object class and representing the likelihood that the one or more images depict an object belonging to the object class. The image processing task may be object detection, where the image processing output identifies one or more regions in the one or more images and, for each region, a likelihood that region depicts an object of interest. As another example, the image processing task can be image segmentation, where the image processing output defines, for each pixel in the one or more images, a respective likelihood for each category in a predetermined set of categories. For example, the set of categories can be foreground and background. As another example, the set of categories can be object classes. As another example, the image processing task can be depth estimation, where the image processing output defines, for each pixel in the one or more images, a respective depth value. As another example, the image processing task can be motion estimation, where the network input includes multiple images, and the image processing outputdefines, for each pixel of one of the input images, a motion of the scene depicted at the pixel between the images in the network input.

[0224] In some implementations, input(s) 2 can be or otherwise represent natural language data. Machine-learned model(s) 1 can process the natural language data to generate an output. As an example, machine-learned model(s) 1 can process the natural language data to generate a language encoding output. As another example, machine-learned model(s) 1 can process the natural language data to generate a latent text embedding output. As another example, machine- learned model(s) 1 can process the natural language data to generate a translation output. As another example, machine-learned model(s) 1 can process the natural language data to generate a classification output. As another example, machine-learned model(s) 1 can process the natural language data to generate a textual segmentation output. As another example, machine- learned model(s) 1 can process the natural language data to generate a semantic intent output. As another example, machine-learned model(s) 1 can process the natural language data to generate an upscaled text or natural language output (e.g., text or natural language data that is higher quality than the input text or natural language, etc.). As another example, machine- learned model(s) 1 can process the natural language data to generate a prediction output (e.g., one or more predicted next portions of natural language content).

[0225] In some implementations, input(s) 2 can be or otherwise represent speech data (e.g., data describing spoken natural language, such as audio data, textual data, etc.). Machine- learned model(s) 1 can process the speech data to generate an output. As an example, machine- learned model(s) 1 can process the speech data to generate a speech recognition output. As another example, machine-learned model(s) 1 can process the speech data to generate a speech translation output. As another example, machine-learned model(s) 1 can process the speech data to generate a latent embedding output. As another example, machine-learned model(s) 1 can process the speech data to generate an encoded speech output (e.g., an encoded and / or compressed representation of the speech data, etc.). As another example, machine-learned model(s) 1 can process the speech data to generate an upscaled speech output (e.g., speech data that is higher quality than the input speech data, etc.). As another example, machine-learned model(s) 1 can process the speech data to generate a textual representation output (e.g., a textual representation of the input speech data, etc.). As another example, machine-learned model(s) 1 can process the speech data to generate a prediction output.

[0226] In some implementations, input(s) 2 can be or otherwise represent latent encoding data (e.g., a latent space representation of an input, etc.). Machine-learned model(s) 1 can process the latent encoding data to generate an output. As an example, machine-learnedmodel(s) 1 can process the latent encoding data to generate a recognition output. As another example, machine-learned model(s) 1 can process the latent encoding data to generate a reconstruction output. As another example, machine-learned model(s) 1 can process the latent encoding data to generate a search output. As another example, machine-learned model(s) 1 can process the latent encoding data to generate a reclustering output. As another example, machine-learned model(s) 1 can process the latent encoding data to generate a prediction output.

[0227] In some implementations, input(s) 2 can be or otherwise represent statistical data. Statistical data can be, represent, or otherwise include data computed and / or calculated from some other data source. Machine-learned model(s) 1 can process the statistical data to generate an output. As an example, machine-learned model(s) 1 can process the statistical data to generate a recognition output. As another example, machine-learned model(s) 1 can process the statistical data to generate a prediction output. As another example, machine-learned model(s) 1 can process the statistical data to generate a classification output. As another example, machine-learned model(s) 1 can process the statistical data to generate a segmentation output. As another example, machine-learned model(s) 1 can process the statistical data to generate a visualization output. As another example, machine-learned model(s) 1 can process the statistical data to generate a diagnostic output.

[0228] In some implementations, input(s) 2 can be or otherwise represent sensor data. Machine-learned model(s) 1 can process the sensor data to generate an output. As an example, machine-learned model(s) 1 can process the sensor data to generate a recognition output. As another example, machine-learned model(s) 1 can process the sensor data to generate a prediction output. As another example, machine-learned model(s) 1 can process the sensor data to generate a classification output. As another example, machine-learned model(s) 1 can process the sensor data to generate a segmentation output. As another example, machine- learned model(s) 1 can process the sensor data to generate a visualization output. As another example, machine-learned model(s) 1 can process the sensor data to generate a diagnostic output. As another example, machine-learned model(s) 1 can process the sensor data to generate a detection output.

[0229] In some implementations, machine-learned model(s) 1 can be configured to perform a task that includes encoding input data for reliable and / or efficient transmission or storage (and / or corresponding decoding). For example, the task may be an audio compression task. The input may include audio data and the output may comprise compressed audio data. In another example, the input includes visual data (e.g. one or more images or videos), theoutput comprises compressed visual data, and the task is a visual data compression task. In another example, the task may comprise generating an embedding for input data (e.g. input audio or visual data). In some cases, the input includes audio data representing a spoken utterance and the task is a speech recognition task. The output may comprise a text output which is mapped to the spoken utterance. In some cases, the task comprises encrypting or decrypting input data. In some cases, the task comprises a microprocessor performance task, such as branch prediction or memory address translation.

[0230] In some implementations, the task is a generative task, and machine-learned model(s) 1 can be configured to output content generated in view of input(s) 2. For instance, input(s) 2 can be or otherwise represent data of one or more modalities that encodes context for generating additional content.

[0231] In some implementations, the task can be a text completion task. Machine-learned model(s) 1 can be configured to process input(s) 2 that represent textual data and to generate output(s) 3 that represent additional textual data that completes a textual sequence that includes input(s) 2. For instance, machine-learned model(s) 1 can be configured to generate output(s) 3 to complete a sentence, paragraph, or portion of text that follows from a portion of text represented by input(s) 2.

[0232] In some implementations, the task can be an instruction following task. Machine- learned model(s) 1 can be configured to process input(s) 2 that represent instructions to perform a function and to generate output(s) 3 that advance a goal of satisfying the instruction function (e.g., at least a step of a multi-step procedure to perform the function). Output(s) 3 can represent data of the same or of a different modality as input(s) 2. For instance, input(s) 2 can represent textual data (e.g., natural language instructions for a task to be performed) and machine-learned model(s) 1 can process input(s) 2 to generate output(s) 3 that represent textual data responsive to the instructions (e.g., natural language responses, programming language responses, machine language responses, etc.). Input(s) 2 can represent image data (e.g., image-based instructions for a task to be performed, optionally accompanied by textual instructions) and machine-learned model(s) 1 can process input(s) 2 to generate output(s) 3 that represent textual data responsive to the instructions (e.g., natural language responses, programming language responses, machine language responses, etc.). One or more output(s) 3 can be iteratively or recursively generated to sequentially process and accomplish steps toward accomplishing the requested functionality. For instance, an initial output can be executed by an external system or be processed by machine-learned model(s) 1 to complete an initial step of performing afunction. Multiple steps can be performed, with a final output being obtained that is responsive to the initial instructions.

[0233] In some implementations, the task can be a question answering task. Machine- learned model(s) 1 can be configured to process input(s) 2 that represent a question to answer and to generate output(s) 3 that advance a goal of returning an answer to the question (e.g., at least a step of a multi-step procedure to perform the function). Output(s) 3 can represent data of the same or of a different modality as input(s) 2. For instance, input(s) 2 can represent textual data (e.g., natural language instructions for a task to be performed) and machine-learned model(s) 1 can process input(s) 2 to generate output(s) 3 that represent textual data responsive to the question (e.g., natural language responses, programming language responses, machine language responses, etc.). Input(s) 2 can represent image data (e.g., image-based instructions for a task to be performed, optionally accompanied by textual instructions) and machine- learned model(s) 1 can process input(s) 2 to generate output(s) 3 that represent textual data responsive to the question (e.g., natural language responses, programming language responses, machine language responses, etc.). One or more output(s) 3 can be iteratively or recursively generated to sequentially process and accomplish steps toward answering the question. For instance, an initial output can be executed by an external system or be processed by machine- learned model(s) 1 to complete an initial step of obtaining an answer to the question (e.g., querying a database, performing a computation, executing a script, etc.). Multiple steps can be performed, with a final output being obtained that is responsive to the question.

[0234] In some implementations, the task can be an image generation task. Machine- learned model(s) 1 can be configured to process input(s) 2 that represent context regarding a desired portion of image content. The context can include text data, image data, audio data, etc. Machine-learned model(s) 1 can be configured to generate output(s) 3 that represent image data that depicts imagery related to the context. For instance, machine-learned model(s) 1 can be configured to generate pixel data of an image. Values for channel(s) associated with the pixels in the pixel data can be selected based on the context (e.g., based on a probability determined based on the context).

[0235] In some implementations, the task can be an audio generation task. Machine- learned model(s) 1 can be configured to process input(s) 2 that represent context regarding a desired portion of audio content. The context can include text data, image data, audio data, etc. Machine-learned model(s) 1 can be configured to generate output(s) 3 that represent audio data related to the context. For instance, machine-learned model(s) 1 can be configured to generate waveform data in the form of an image (e.g., a spectrogram). Values for channel(s) associatedwith pixels of the image can be selected based on the context. Machine-learned model(s) 1 can be configured to generate waveform data in the form of a sequence of discrete samples of a continuous waveform. Values of the sequence can be selected based on the context (e.g., based on a probability determined based on the context).

[0236] In some implementations, the task can be a data generation task. Machine-learned model(s) 1 can be configured to process input(s) 2 that represent context regarding a desired portion of data (e.g., data from various data domains, such as sensor data, image data, multimodal data, statistical data, etc.). The desired data can be, for instance, synthetic data for training other machine-learned models. The context can include arbitrary data type(s). Machine-learned model(s) 1 can be configured to generate output(s) 3 that represent data that aligns with the desired data. For instance, machine-learned model(s) 1 can be configured to generate data values for populating a dataset. Values for the data object(s) can be selected based on the context (e.g., based on a probability determined based on the context).Additional Disclosure

[0237] The technology discussed herein makes reference to servers, databases, software applications, and other computer-based systems, as well as actions taken and information sent to and from such systems. The inherent flexibility of computer-based systems allows for a great variety of possible configurations, combinations, and divisions of tasks and functionality between and among components. For instance, processes discussed herein can be implemented using a single device or component or multiple devices or components working in combination. Databases and applications can be implemented on a single system or distributed across multiple systems. Distributed components can operate sequentially or in parallel.

[0238] While the present subj ect matter has been described in detail with respect to various specific example embodiments thereof, each example is provided by way of explanation, not limitation of the disclosure. Those skilled in the art, upon attaining an understanding of the foregoing, can readily produce alterations to, variations of, and equivalents to such embodiments. Accordingly, the subject disclosure does not preclude inclusion of such modifications, variations and / or additions to the present subject matter as would be readily apparent to one of ordinary skill in the art. For instance, features illustrated or described as part of one embodiment can be used with another embodiment to yield a still further embodiment. Thus, it is intended that the present disclosure covers such alterations, variations, and equivalents.

Claims

WHAT IS CLAIMED IS:

1. A computing system for maintaining a knowledge graph of a user, comprising: one or more processors; and one or more non -transitory computer-readable media that collectively store: a database storing the knowledge graph having a plurality of structured data entries; a machine-learned model, wherein the machine-learned model is configured to determine an insight; and instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising: obtaining, from a user device, a first user entry; processing, using the machine-learned model, the first user entry with a prompt to generate a first structured data entry; processing the first structured data entry and the knowledge graph to determine the first insight; processing, using the machine-learned model, the first insight to determine a layout of a graphical user interface that is presented on the user device; and causing a presentation of the first insight on the graphical user interface of the user device.

2. The computing system of claim 1, wherein the knowledge graph is updated by including the first structured data entry in the plurality of structured data entries.

3. The computing system of claim 1, wherein the first user entry includes image data and user speech data.

4. The computing system of claim 1, wherein the first user entry includes video data and user speech data.

5. The computing system of claim 1, wherein the first user entry is an image from the user device, the operations further comprising: processing, using an image classification model, the image to determine a classification type of an object in the image,wherein the first structured data entry is further determined based on the classification type of the image in the model, and wherein the layout of the graphical user interface is determined based on the classification type of the image.

6. The computing system of claim 5, wherein the classification type of the object is a first classification, the operations further comprising: obtaining, via the graphical user interface, a second user entry modifying the classification type of the object in the image to a second classification; updating the classification type of the object in the image based on the second user entry; and presenting the object in the image with the second classification in the graphical user interface.

7. The computing system of claim 6, wherein the classification type is a size associated with the object, and wherein the user input changes the size associated with the object.

8. The computing system of claim 1, the operations further comprising: obtaining, from a nutrition database, nutrition data associated with an object in the first user entry, and wherein the first structured data entry is generated based on the nutrition data associated with the object; presenting the first structured data entry in the graphical user interface; receiving a second user entry modifying the first structured data entry; and modifying the first structured data entry based on the second user entry, wherein the knowledge graph is updated by including the first structured data entry in the plurality of structured data entries.

9. The computing system of claim 1, the operations further comprising: obtaining location data from the user device, wherein the first structured data entry is generated based on the location data associated with the object; and processing, using the machine-learned model, the first structured data, the knowledge graph, and the location data to determine the first insight.

10. The computing system of claim 1, the operations further comprising: determining the layout of the graphical user interface based on a classification type of the first structured data entry.

11. The computing system of claim 1, the operations further comprising: determining the layout of the graphical user interface based on a category associated with the first insight.

12. The computing system of claim 1, the operations further comprising: determining the layout of the graphical user interface further based on a value of an attribute of the first structured data entry.

13. The computing system of claim 1, the operations further comprising: analyzing the knowledge graph to determine a patterned-input received at a specific time; and prompting for a second user entry after determining that the patterned-input has not been received by the specific time.

14. The computing system of claim 1, the operations further comprising: obtaining, from a sensor device, sensor data associated with a user of the user device; processing, using the machine-learned model, the sensor data to generate a second structured data entry; and storing the second structured data entry in the knowledge graph.

15. The computing system of claim 14, the operations further comprising: obtaining, from a sleep pattern database, a first sleep pattern associated with the sensor data, and wherein the first structured data entry is generated based on the first sleep pattern.

16. The computing system of claim 1, wherein the insight is a new habit of the user, and wherein the machine-learned model tracks the new habit of the user.

17. A computer-implemented method for maintaining a knowledge graph having a plurality of structured data entries, comprising:obtaining, from a user device, a first user entry; processing, using the machine-learned model, the first user entry with a prompt to generate a first structured data entry; processing the first structured data entry and a knowledge graph to determine the first insight, the knowledge graph having a plurality of structured data entries; processing, using a machine-learned model, the first insight to determine a layout of a graphical user interface that is presented on the user device; and causing a presentation of the first insight on the graphical user interface of the user device.

18. The computer-implemented method of claim 17, further comprising: obtaining an image from the user device; and processing, using an image classification model, the image to determine a classification type of an object in the image, wherein the first structured data entry is further determined based on the classification type of the object, and wherein the layout is further determined based on the classification type of the object.

19. The computer-implemented method of claim 17, further comprising: analyzing the knowledge graph to determine a patterned-input received at a specific time; and prompting for a second user entry after determining that the patterned-input has not been received by the specific time.

20. One or more non-transitory computer-readable media storing instructions that are executable by a computing system to perform operations, the operations comprising: obtaining, from a user device, a first user entry; processing, using the machine-learned model, the first user entry with a prompt to generate a first structured data entry; processing the first structured data entry and a knowledge graph to determine the first insight, the knowledge graph having a plurality of structured data entries; processing, using a machine-learned model, the first insight to determine a layout of a graphical user interface that is presented on the user device; andcausing a presentation of the first insight on the graphical user interface of the user device.

Images