12 results about "Memory consolidation" patented technology

The invention discloses a dynamic sleep memory intervention method and system based on an ultrathin eyeshade, the system comprises an ultrathin eyeshade body and a cloud platform, and the eyeshade integrates electroencephalogram acquisition, memory content playing, master control and communication modules. The electroencephalogram acquisition module adopts an eight-channel super-soft spandex conductive electrode and a noise reduction unit combining adaptive notch filtering with wavelet threshold filtering; the memory playing module is a hemispherical dome miniature bone conduction loudspeaker, and silent leakage outside 30cm is realized; a dynamic sleep memory dry prediction algorithm is built in the main control module, and playing parameters can be adjusted according to delta wave intensity; the cloud stores personalized memory content and intervention records, and push content can be intelligently updated. The method comprises the steps of equipment starting initialization, electroencephalogram collection and noise reduction, deep sleep recognition, dynamic playing and data encryption synchronous closing, the problems that existing equipment is heavy and intervention is rigid are solved, non-inductive wearing, accurate intervention and privacy protection are achieved, and the method is suitable for sleep memory consolidation.

The invention discloses a memory enhancement-based large language model lifelong alignment method. The method comprises the steps of focus preference optimization and short-time to long-time memory consolidation. The focus preference optimization adaptively focuses the learning focus on a new preference sample or a preference sample with an uncertain model through an improved preference learning loss function, meanwhile, the updating amplitude of fully learned knowledge is reduced, and a historical alignment result is protected while new preference is learned; the short-term to long-term memory consolidation is used for simulating a human memory mechanism, denoising is performed on short-term parameter update through singular value decomposition, an update part conflicting with past knowledge is identified and suppressed by projecting to a historical knowledge subspace, and finally refined conflict-free knowledge is integrated into long-term parameters of a model. According to the method, knowledge of the model is effectively accumulated and reserved in continuous and diversified alignment tasks, catastrophic forgetting is remarkably inhibited, and the stability, reliability and alignment consistency of the model in a dynamic and long-term deployment environment are improved.

The present application relates to the technical field of intelligent health management, and particularly relates to a multi-modal intelligent health management method and system based on continuous memory, comprising: multi-modal preprocessing of voice, text, images and sensor data input by a user to obtain multi-modal health data; adopting a hierarchical continuous memory storage strategy to obtain a long-term memory item set through three-layer memory architecture step-by-step compression processing; constructing a health memory graph network based on a memory consolidation and association integration strategy; adopting a global health portrait incremental update and memory portrait driven parameter mapping strategy to map a health portrait into a role parameter vector; adopting a memory retrieval and knowledge graph enhanced reasoning strategy to combine historical memory recall, knowledge graph reasoning and drug information management results to generate and output a health management response. The present application realizes the organic unification of continuous memory accumulation, personalized accompaniment and intelligent health reasoning for a health management scene.

This invention relates to the fields of biomedical signal processing and brain-computer interface technology, specifically a short-sleep cognitive recovery system and method based on lightweight deep learning. The system includes: an EEG signal acquisition module, a signal preprocessing module, a sleep staging inference module, a cognitive state assessment module, and a closed-loop acoustic intervention module. The EEG acquisition module employs a low sampling rate of 64Hz to ensure a target frequency band signal retention rate of ≥95%. The sleep staging inference module uses a dual-stream discriminant architecture, with the XGBoost model using SHAP to select Top-20 features, and the SleepTransformer model containing a 4-layer encoder and 8 attention heads. The cognitive state assessment module calculates the functional connectivity strength of specific brain regions based on the PLV / wPLI index, with preset scientific thresholds. The closed-loop acoustic intervention module dynamically adjusts acoustic parameters, with a response latency ≤100ms. This invention solves the problem of low staging accuracy under low sampling rates in portable devices, achieving precise closed-loop intervention based on neural mechanisms, and significantly improving the efficiency of short-sleep memory consolidation.

The invention provides a multi-expert collaborative knowledge tracking method and system for cognitive strategy perception, and relates to the technical field of knowledge tracking, and the method comprises the steps: building a memory consolidation expert module through combining a knowledge obtaining process with a forgetting gate and absorptivity parameters, building an attention regulation expert module through combining attention distribution of knowledge points with a time attention mechanism, and building a memory consolidation expert module; creating a structure adaptation expert module based on a correlation evolution process between knowledge points; generating a global memory matrix by combining an attention mechanism based on the long-term memory of the historical cognitive state of the student and the current interaction embedded information, and determining a corresponding expert weight; performing weighted fusion on knowledge state parameters generated by the memory consolidation expert module, the attention regulation and control expert module and the structure adaptation expert module on the basis of expert weights, splicing the obtained fused knowledge state parameters and feature embedding vectors of test questions to be predicted, and inputting the spliced information to a preset feedforward neural network, and obtaining the answer accuracy corresponding to the test question to be predicted.

This invention provides a method for calculating the sleep memory consolidation index based on electroencephalogram (EEG) signals, comprising the following specific steps: Step 1: Testing the subject's pre-sleep memory test value and post-sleep memory test value, and collecting single-lead EEG data and PSG multi-lead EEG data during the subject's sleep; Step 2: Segmenting the single-lead EEG data; Step 3: Performing sleep staging on the single-lead EEG data; Step 4: Calculating the total number of spindle waves N. sp Spindle wave ratio R sp Average amplitude of spindle wave (Amp) avg Step 5: Establish the calculation formula for the sleep memory consolidation index; Step 6: Determine the characteristic coefficients in the calculation formula through multiple linear regression. This invention provides a method for calculating the sleep memory consolidation index, which can intuitively characterize the level of memory consolidation after sleep in subjects.

The invention relates to the technical field of biomedical signal processing and intelligent wearable equipment, and discloses a real-time electroencephalogram-based closed-loop targeted memory reactivation system and method.The system comprises a wearable equipment body, an electroencephalogram sensing module, a central processing unit and an audio playing module, and the electroencephalogram sensing module collects electroencephalogram signals of a user in real time; the central processing unit is used for processing the electroencephalogram signals, including sleep staging, feature recognition and decision triggering; the audio playing module plays the memory-related audio clues under specific conditions. The system accurately locates a'golden window period 'of memory consolidation by identifying a time phase coupling relationship between slow wave oscillation and sleep spindle waves in sleep to realize targeted intervention of closed-loop control, and the system further comprises a non-stress period setting module, a self-adaptive adjusting module and a multi-mode physiological sensing module, realizes full-automatic, accurate and interference-free memory enhancement, and has a good application prospect. Compared with the prior art, the method has the remarkable advantages of accuracy, no interference and high efficiency.

This invention relates to the field of intelligent health management technology, and more particularly to a multimodal intelligent health management method and system based on persistent memory. The method includes: performing multimodal preprocessing on user-input voice, text, image, and sensor data to obtain multimodal health data; employing a hierarchical persistent memory storage strategy and a three-layer memory architecture for progressive compression processing to obtain a set of long-term memory entries; constructing a health memory graph network based on memory consolidation and association integration strategies; mapping the health profile to role parameter vectors using a global health profile incremental update and memory profile-driven parameter mapping strategy; and generating and outputting a health management response by employing memory retrieval and knowledge graph-enhanced reasoning strategies, combined with historical memory retrieval, knowledge graph reasoning, and drug information management results. This invention achieves the organic unity of persistent memory accumulation, personalized companionship, and intelligent health reasoning for health management scenarios.

The invention belongs to the field of nerve regulation and control technology and biomedical engineering, and particularly relates to a closed-loop transcranial electrical stimulation system and method for enhancing sleep memory consolidation, and the system comprises a wearable device main body, a multi-channel electrode array module, a biological signal collection unit, a central processing unit, a precise constant current source module and a safety monitoring module. According to the system, the instantaneous phase of slow wave oscillation in electroencephalogram signals is tracked in real time, micro-current stimulation synchronous with endogenous neural oscillation is applied to an accurate window in the slow wave rising period, multi-dimensional regulation and control over slow waves are achieved, a predictive phase tracking algorithm is adopted by the system, and stimulation phase errors are controlled within the radian of pi / 8; natural propagation of slow waves is simulated through a forehead-temporal stimulation mode and a spatial orientation stimulation mode; clinical verification shows that the system can enhance the slow wave amplitude by 32%, prolong the duration by 26%, improve the density by 20% and improve various memory task expressions by 20%-28%.

A synaptic array includes a plurality of Fowler-Nordheim (FN) synapses. Each FN synapse connected to at least one other FN synapse of the plurality of FN synapses to form a network. Each FN synapse includes a pair of FN tunneling devices each including a floating gate. Each FN synapse is operable to store a synaptic weight as a differential voltage across the floating gates of its FN tunneling devices and to implement synaptic memory consolidation.

This invention discloses a memory management method and device for urban rail intelligent agents. The method extracts multi-dimensional feature vectors from memory fragments and constructs a dynamic decay prediction model for memory traces. It predicts the future recall probability at set time intervals and calculates the expected value based on a value coefficient. Memory levels are categorized according to thresholds to achieve hierarchical storage and retrieval priority allocation. The recall probability is monitored in real time, and memory consolidation is actively triggered when conditions are met. When a memory is recalled, reconsolidation is performed based on information increments, updating the trace strength and feature vectors. The error is calculated based on the prediction results and actual recall, a dataset is constructed, and the model is adaptively optimized. This invention solves the problems of existing technologies, such as lack of scientific basis for resource allocation, failure to simulate cognitive patterns, and lack of adaptive model optimization.

A memory consolidation controller includes: an acquisition unit configured to acquire brain activity information indicating a state of brain activity of a subject; a determination unit configured to determine a sleep level indicating a depth of sleep of the subject and determine an activity state in a predetermined region of the subject's brain based on the acquired brain activity information; and a reproduction control unit configured to cause a reproduction device to reproduce a content that stimulates an auditory sense of the subject based on a determination result of the determination unit.