36 results about "Information bottleneck method" patented technology

A method for reengineering of a business process. The method includes extracting baseline information requirements of at least one business process, wherein the step of extracting includes interviewing a process owner of each of the sub-processes of the business process based on a predetermined questionnaire and quantifying information requirements of each of the sub-processes of the business process based on the questionnaire. The method also includes displaying the baseline information requirements in a predetermined matrix structure and prioritizing opportunities for removing information bottlenecks in the at least one business process using the predetermined matrix structure.

The invention relates to a text abstraction method based on hierarchical interaction attention, and belongs to the technical field of natural language processing. According to the method, the generation of the abstract is guided by extracting the feature information of different levels of the encoder through the level interaction attention. Meanwhile, in order to avoid the problem of information redundancy caused by introduction of different levels of features, data noise is compressed by utilizing a variational information bottleneck. Aiming at the generative text abstract, under an attention-based encoding and decoding framework, the multi-layer context information of the encoder is extracted through an attention mechanism to guide the decoding process, and meanwhile, the variational information bottleneck is introduced to constrain the information, so that the quality of the generative text abstract is improved. Experimental results show that the method can significantly improve theperformance of the encoding and decoding framework in a generative abstract task.

The invention discloses an identification method based on a queuing detector information bottleneck state. The identification method provided by the invention comprises the following steps of: adopting the detection information of a queuing detector as a foundation; and carrying out real-time discrimination on a road section transportation state by taking the continuously occurring number of occupancies of rolling time exceeding a threshold as a discrimination index, and distinguishing the position of a detector as a bottleneck state when a queue length is greater than or equal to the distance of the queuing detector and the stop line. The identification method provided by the invention adopts the rolling time occupancy as the discrimination index of the bottleneck state, and improves the discrimination real-time property of the bottleneck state.

A system for improved localization of image forgery. The system generates a variational information bottleneck objective function and works with input image patches to implement an encoder-decoder architecture. The encoder-decoder architecture controls an information flow between the input image patches and a representation layer. The system utilizes information bottleneck to learn useful residual noise patterns and ignore semantic content present in each input image patch. The system trains a neural network to learn a representation indicative of a statistical fingerprint of a source camera model from each input image patch while excluding semantic content thereof. The system can determine a splicing manipulation localization by the trained neural network.

Some embodiments of the invention provide a novel method for training a quantized machine-trained network. Some embodiments provide a method of scaling a feature map of a pre-trained floating-point neural network in order to match the range of output values provided by quantized activations in a quantized neural network. A quantization function is modified, in some embodiments, to be differentiable to fix the mismatch between the loss function computed in forward propagation and the loss gradient used in backward propagation. Variational information bottleneck, in some embodiments, is incorporated to train the network to be insensitive to multiplicative noise applied to each channel. In some embodiments, channels that finish training with large noise, for example, exceeding 100%, are pruned.

The invention discloses a small sample image classification method based on a memory mechanism and a graph neural network, which is characterized in that a small sample model is helped to perform reasoning prediction by means of learned conceptual knowledge, and specifically comprises three stages of pre-training, meta-training and meta-testing, wherein the pre-training takes the trained feature extractor and classifier as initialization weights of an encoder and a memory bank; the meta-training is characterized in that features of samples of a support set and a query set are extracted through an encoder, related information of each class is mined from a memory bank to serve as meta-knowledge, and similarity between task related nodes and the meta-knowledge is propagated through a graph neural network; and the meta-test obtains a classification result through task related nodes and meta-knowledge nodes. Compared with the prior art, the method has the advantages that a human recognition process is used for reference, a memory graph augmentation network based on information bottleneck is used, well-learned conceptual knowledge is used, the model is helped to conduct reasoning prediction, the method is simple and convenient, practicability is high, and certain application and popularization prospects are achieved.

The invention relates to a transformer fault diagnosis method based on information bottleneck, and belongs to the technical field of transformer fault diagnosis. The method is that the information bottleneck method is adopted to cluster data to obtain a result cluster with the maximum similarity in the cluster; a cluster label of a sample to be detected is determined by voting training samples in the cluster so as to determine the fault class. With the adoption of the method, the classification result is superior to that of the rule-based Duval triangle method, the mode recognizing based BPNN algorithm and the mode recognizing based Bayes method; the diagnosis accuracy can be effectively increased during diagnosing the transformer fault.

Some embodiments of the invention provide a novel method for training a quantized machine-trained network. Some embodiments provide a method of scaling a feature map of a pre-trained floating-point neural network in order to match the range of output values provided by quantized activations in a quantized neural network. A quantization function is modified, in some embodiments, to be differentiable to fix the mismatch between the loss function computed in forward propagation and the loss gradient used in backward propagation. Variational information bottleneck, in some embodiments, is incorporated to train the network to be insensitive to multiplicative noise applied to each channel. In some embodiments, channels that finish training with large noise, for example, exceeding 100%, are pruned.

The invention discloses a probability domain generalization learning method based on meta-learning, belongs to the field of meta-learning, and aims to combine a meta-learning thought into domain generalization for the first time and solve the problem that parameters are linearly increased along with increase of the number of source domains in domain generalization by utilizing a meta-learning framework. The variational information bottleneck idea is combined into meta-learning and domain generalization for the first time, so that the generalization ability of the patent can be further improved; according to the method, the problem that parameters linearly increase along with the number of source domains can be solved through meta-learning; according to the technical scheme, the probabilitydomain generalization learning method based on meta-learning is formed by combining the variational thought with the information bottleneck and fusing the variational thought and the information bottleneck into a unified probability framework, and the brand-new and effective probability domain generalization learning method based on meta-learning can be formed by combining the variational thoughtwith the information bottleneck.

The invention discloses an unsupervised domain adaptive method for beneficial feature alignment under class conditions, and the method comprises the steps of calculating all source domain images and target domain images, and obtaining a pseudo tag of the target domain image; decoupling the pseudo labels of the source domain image and the target domain image through variational information bottlenecks, filtering features irrelevant to tasks out, and obtaining beneficial and migratable features; estimating intra-class difference and inter-class spacing by using a conditional slice Wharisstein distance, minimizing the intra-class difference and maximizing the inter-class spacing in a cross-domain manner, reducing class-level distribution difference between a source domain and a target domain, and obtaining domain-invariant discriminant features. According to the method, decoupling of a source domain and a target domain can be achieved, class-level information is embedded into the slice Weiisstein distance, beneficial feature alignment is achieved, and meanwhile beneficial feature decoupling and class condition feature alignment are achieved so as to promote better domain adaptation.

The invention discloses a gesture recognition method and device based on random variation information, equipment and a medium, and the method comprises the steps: carrying out the feature extraction of obtained to-be-detected data, and obtaining the feature data in different modes; obtaining a variational information bottleneck; compressing random local information in the feature data through the variational information bottleneck, and reserving residual local information in the feature data to obtain a fusion result of the feature data in different modes; and inputting the fusion results in the different modes into a preset model, and outputting a gesture recognition result. According to the invention, the relevance among multiple modes can be fully utilized, the interference noise is reduced, the recognition accuracy is further improved, and the method can be widely applied to the technical field of image processing.

The invention provides a voiceprint recognition method based on variational information bottleneck and a system thereof, and solves the problems of poor robustness and low discrimination of speaker embedding extracted by an existing voiceprint recognition model. The method comprises the following steps: firstly, providing a feature extraction network consisting of VovNet and an ultra-lightweight subspace attention mechanism (ULSAM), wherein the feature extraction network is used for extracting multi-scale and multi-frequency frame-level speaker information; and then introducing a variational information bottleneck as a regularization method, further compressing the feature vector of the speaker, removing information irrelevant to the speaker, and only keeping information relevant to identification of the identity of the speaker, so that embedding of the finally extracted speaker is more robust. Compared with an existing voiceprint recognition technology, the voiceprint recognition method improves the recognition accuracy of voiceprint recognition under the noise background, and enables the voiceprint recognition technology to be more suitable for actual life scenes.

The invention belongs to the field of natural language processing and sentiment analysis, and particularly relates to a real estate market analysis method and device based on deep transfer learning, and equipment, and the real estate market analysis method comprises the steps: collecting real estate network public opinion data, and carrying out the preprocessing of the public opinion data; constructing a deep multi-channel neural network integrated with a variational information bottleneck; pre-training the network by using a large amount of annotation data in the related field; finely adjusting a pre-established network by using a small amount of marked public opinion data in a transfer learning mode; and performing emotional tendency analysis on unlabeled real estate public opinion dataon the migrated network, and obtaining a final real estate market emotional analysis result. According to the real estate market analysis method, deep migration learning and real estate network publicopinions are combined, and real estate market emotion can be accurately analyzed, so that reference and guidance are provided for policy making of related departments, decision deployment of companies and enterprises and investment planning of individual consumers.

The invention relates to the field of natural language processing, and discloses a text classification method and device, a medium and electronic equipment. The method comprises: obtaining target text data; inputting the target text data into a pre-trained text classification model; outputting compressed sentence representation information corresponding to the target text data and an expected value corresponding to the compressed sentence representation information through a variational information bottleneck processing layer; outputting classification prediction information through a classification module according to the compressed sentence representation information received from the variational information bottleneck processing layer; and generating and outputting a classification label corresponding to the target text data through a classification label generation layer according to the classification prediction information received from a classification module and an expected value corresponding to the compressed sentence representation information received from the variational information bottleneck processing layer. According to the method, the occurrence of an over-fitting phenomenon is reduced, and the popularization and application range of the pre-training model is expanded.

The invention relates to a reinforcement learning battle game AI training method based on an information bottleneck theory. The method comprises the following steps of 1) initializing an AI training model; 2) performing decision interaction in a simulation environment through the game AI to obtain a sample training batch data set; 3) iteratively training an AI training model by adopting a reinforcement learning algorithm according to a sample training batch data set obtained by interaction between the game AI and the environment, and storing parameters of the AI training model in stages; and (4) fixing part of the stored parameters of the AI training models at different stages, training the remaining parameters again by using a reinforcement learning algorithm to perform fine adjustment to obtain the final AI training models of different levels of AIs, and generating an AI file of the battle game. Compared with the prior art, the method has advantages of high sampling efficiency, high training speed, high test flexibility, AI grading and the like.

The invention provides a three-dimensional fuzzy clustering method based on an information bottleneck theory. The three-dimensional fuzzy clustering method provided by the invention can be used for clustering the user-searching-document three-dimensional data. After the three-dimensional fuzzy clustering method is used, the simultaneous clustering of three dimensions can be realized for a three-dimensional contingency table, and the clustering accurate rate can be effectively enhanced on the basis of the distance between an information bottleneck theory measurement document and a cluster centroid.

The invention belongs to the technical field of medical image processing, and provides a breast cancer image information bottleneck multi-task classification and segmentation method and system. The method comprises the following steps: acquiring a plurality of breast images of contrast enhanced X-ray photography and corresponding benign and malignant categories and lump position pixel-level labels; performing preprocessing operation on each acquired breast image of the contrast enhanced X-ray photography; adopting a multi-task network to extract multi-task shared representation for each preprocessed mammary gland image; inputting the shared representation to a classification encoder to obtain an encoding tensor of an intermediate layer, and then sending the intermediate encoding tensor to an information bottleneck attribution module for feature compression and benign and malignant classification to obtain a classification task tensor; adapting the shared representation to a segmentation network to obtain a segmentation task tensor; and carrying out feature fusion on the classification task tensor and the segmentation task tensor to segment the focus.

The invention discloses a deep reinforcement learning model robustness enhancement method based on information bottleneck. According to the deep reinforcement learning model robustness enhancement method based on the information bottleneck, state information in deep reinforcement learning is limited by setting the information bottleneck, the state information in a transfer tuple is encoded through an encoder, firstly, the state observed in the environment is encoded, encoding the data, inputting the encoded data into a strategy network, interacting with the environment according to the action of the strategy network to obtain the state of the next round, encoding the state, and continuously interacting with the environment to realize the training of the strategy network. According to the deep reinforcement learning model robustness enhancement method based on the information bottleneck disclosed by the invention, a strategy obtained by training still has good performance on an original task, and the influence of adversarial attacks can be resisted; a proportionality coefficient in a regular term is set by adopting an annealing thought, so that a stable training process is achieved, and a strategy obtained by training still has excellent performance in a normal task.

The invention provides a communication-sensitive multi-agent cooperation method, which is applied to the technical field of Internet of Vehicles communication. The method comprises the following steps that: an intelligent agent encodes a locally observed message into a hidden state vector by using an information filtering module, i.e., compresses the filtered message, scores the message by using a shared medium access control module before sending the message, ranks the message value by an edge computing node according to the score, and sends the message value to an edge computing node; sending confirmation information to the intelligent agent with the topK message value, and sending the message after the intelligent agent receives the confirmation; the edge computing node receives all messages, extracts effective information by using a message distillation module, and gathers and distributes the effective information to each agent; in the process of summarizing the agent messages, a graph convolution summarizing mode based on a graph information bottleneck is utilized, topological structure information of the agents is reserved, and the summarized messages are distilled. According to the method, transmission of redundant and invalid information among the intelligent agents is reduced, communication resources are saved, and the decision-making ability and cooperation efficiency of the multiple intelligent agents are improved.

The invention discloses a safety assessment framework building method for deep learning bypass analysis. The method comprises the steps that a secret key information quantum framework based on modeling output and a deep learning sub-framework composed of cross entropy and a loss function are included; and the bypass analysis sub-framework consists of a condition guessing entropy, a derivative guessing entropy and a success rate. According to information bottleneck theoretical analysis, perception information from the last hidden layer to the output layer of the DNN is calculated through probability distribution to serve as a quantitative evaluation index of the DLSCA, and experiments prove that the index is more accurate in measurement compared with an existing quantitative evaluation index and has lower requirements for the order of magnitude of data.

The invention relates to the field of artificial intelligence, and particularly discloses a voice classification model training method and device, equipment and a storage medium, and the method comprises the steps: obtaining sample data which comprises a sample voice and a sample label corresponding to the sample voice; preprocessing the sample voice to obtain a vector matrix corresponding to the sample voice; inputting the vector matrix into a variational information bottleneck processing network of an initial voice classification model to obtain sentence representation corresponding to the sample voice; and inputting the sentence representation into a classification network of the initial voice classification model to obtain a prediction label, and then performing iterative training on the initial voice classification model according to the sample label and the prediction label to obtain a trained voice classification model. A variational information bottleneck processing network is added into a voice classification model, so that sample information is compressed, useless information is inhibited, and an over-fitting phenomenon is reduced.

The invention provides an information bottleneck-based possibility fuzzy joint clustering method. An information bottleneck principle is introduced, mutual information loss of the information bottleneck is adopted as a distance measurement mode, and advantages of possibility clustering and fuzzy joint clustering are combined at the same time to standardize the centroid in the clustering process. After the method is adopted, the sensitivity to initial values can be reduced, the robustness and the anti-noise ability are better, the clustering precision is higher, and a more distinct fuzzy partition result is generated.

The invention provides a microblog hot topic detection method and apparatus and relates to the technical field of network information digging. Difficulty for detection of hot topics of highly-sparse microblog texts can be eliminated. The method comprises steps of extracting meaningful strings in an information flow and conducting heat analysis to the meaning string to screen hotspot features, combining hotspot features according to information bottleneck theory, calculating combining degree indication parameters after the combination, and determining the hot topic based on the combining degree indication parameters.

The invention discloses an identification method based on a queuing detector information bottleneck state. The identification method provided by the invention comprises the following steps of: adopting the detection information of a queuing detector as a foundation; and carrying out real-time discrimination on a road section transportation state by taking the continuously occurring number of occupancies of rolling time exceeding a threshold as a discrimination index, and distinguishing the position of a detector as a bottleneck state when a queue length is greater than or equal to the distance of the queuing detector and the stop line. The identification method provided by the invention adopts the rolling time occupancy as the discrimination index of the bottleneck state, and improves the discrimination real-time property of the bottleneck state.

The invention designs an online learning input degree identification method and system based on deep learning, and the method comprises the steps: firstly, carrying out the student face detection through YOLOv4 in order to guarantee that an image is not affected by an irrelevant background; secondly, aiming at the problems that the VGG16 network is huge in parameter quantity, time-consuming in training and the like, an improved VGG16 model is provided, and meanwhile, in the model training process, a depth deterministic information bottleneck method DIB is adopted to make up for the deficiency of a traditional loss function so as to obtain relatively compact feature expression, reduce generalization errors and improve the universality and stability of the model, and achieving precise identification of the learning input degree in a complex online learning scene; and finally, through comparison and analysis with various methods such as traditional machine learning and other deep learning, the effectiveness of the method is verified.

The invention discloses an intelligent mammary gland contrast enhancement photography classification system based on information bottleneck. The system comprises: an image acquisition module which for acquiring mammary gland contrast enhancement photography images and corresponding benign and malignant category labels; an image preprocessing module which carries out preprocessing operation on the mammary gland contrast enhancement photographic image; a feature extraction module which extracts image features from the preprocessed mammary gland contrast enhanced photographic image; an information bottleneck module which calculates image features, benign and malignant category labels and mutual information between input images based on an information bottleneck method so as to obtain a network loss function; and an image classification module which trains a classification model based on the image features and the loss function, and obtains a benign and malignant classification result of the to-be-classified image based on the trained classification model. The problem of accurate classification of mammary gland contrast enhancement photography based on artificial intelligence is solved.

The invention relates to the technical field of artificial intelligence, and provides a named entity recognition method, device and equipment and a computer readable storage medium, and the named entity recognition method comprises the steps: obtaining a pre-trained named entity recognition model, obtaining a to-be-recognized first statement, and inputting the to-be-recognized first statement into the named entity recognition model, performing the following named entity recognition processing by using the mission name entity recognition model: performing word segmentation processing on the first statement to obtain a second statement comprising a plurality of segmented words; performing feature extraction on the plurality of split words to obtain a plurality of word embedding feature vectors; processing the second statement according to the plurality of word embedding feature vectors to obtain a plurality of cross-domain information features; processing the plurality of cross-domain information features through an information bottleneck layer to obtain a plurality of information bottleneck features; the classification function is adopted to classify and recognize the multiple information bottleneck features, the corresponding named entity category is determined, the unregistered words in the named entity can be better recognized, and the named entity recognition accuracy is improved.