Neural Network Memory Computing System for Sense-Making
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current deep neural network (DNN) technology is efficient for name recognition but inefficient for sense making, requiring large amounts of training data to grasp meaning from various modality inputs, necessitating a technology for learning and inferring meaning from complex inputs.
Innovation Solution
A neural network memory computing system and method that includes processors for learning sense-making processes using multimodal training data, embedding data into vectors, generating and updating training sets, and using deep neural networks to classify and output sense-making results, enabling the system to grasp meanings of complex inputs across multiple modalities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current deep neural network technology is used for sense making, then name recognition efficiency is maintained, but sense-making capability deteriorates due to insufficient training data
Solution Approach 1:
The system performs preliminary actions by automatically generating sense-making training data before actual sense-making tasks. The data generation module creates synthetic training datasets with predefined meanings and associations, allowing the neural network to learn sense-making capabilities without requiring extensive manual training data collection
Solution Approach 2:
The system serves itself by automatically generating its own training data. The data generation module uses the neural network's existing knowledge and predefined meaning information to create training examples, eliminating the need for external data sources or manual annotation processes
2Measurement precision
If large amounts of training data are collected for sense making, then sense-making accuracy is improved, but system complexity and data processing overhead increase
Solution Approach 1:
The system changes parameters by using predefined meaning information and semantic relationships instead of relying on large volumes of raw training data. The data generation module transforms simple meaning definitions into complex training examples, achieving high sense-making accuracy with minimal input data
Solution Approach 2:
The system introduces an intermediary layer of meaning information that bridges input data and output interpretations. This intermediary layer contains predefined semantic relationships and associations that guide the neural network's sense-making process, reducing the need for extensive training data
Data Source
AI summary
Provided are a neural network memory computing system and method. The neural network memory computing system includes a first processor configured to learn a sense-making process on the basis of sense-making multimodal training data stored in a database, receive multiple modalities, and output a sense-making result on the basis of results of the learning, and a second processor configured to generate a sense-making training set for the first processor to increase knowledge for learning a sense-making process and provide the generated sense-making training set to the first processor.