Search

Recent experimental evidence suggests that oscillatory activity plays a pivotal role in the maintenance of information in working memory, both in rodents and humans. In particular, cross-frequency coupling between theta and gamma oscillations has been suggested as a core mechanism for multi-item memory. The aim of this work is to present an original neural network model, based on oscillating neural masses, to investigate mechanisms at the basis of working memory in different conditions. We show that this model, with different synapse values, can be used to address different problems, such as the reconstruction of an item from partial information, the maintenance of multiple items simultaneously in memory, without any sequential order, and the reconstruction of an ordered sequence starting from an initial cue. The model consists of four interconnected layers; synapses are trained using Hebbian and anti-Hebbian mechanisms, in order to synchronize features in the same items, and desynchronize features in different items. Simulations show that the trained network is able to desynchronize up to nine items without a fixed order using the gamma rhythm. Moreover, the network can replicate a sequence of items using a gamma rhythm nested inside a theta rhythm. The reduction in some parameters, mainly concerning the strength of GABAergic synapses, induce memory alterations which mimic neurological deficits. Finally, the network, isolated from the external environment ("imagination phase") and stimulated with high uniform noise, can randomly recover sequences previously learned, and link them together by exploiting the similarity among items. [ABSTRACT FROM AUTHOR]

This article describes navigation data of 14 month-old APPPS1 and C57Bl6 in the Starmaze task. These data were acquired as positive controls of memory deficit in a model of the familial form of Alzheimers's disease (see Schmitt et al., Flexibility as a marker of early cognitive decline in humanized Apolipoprotein E ε4 (ApoE4) mice, Neurobiol Aging, 2021). They were acquired in a reduced version of the Starmaze environment and accompanied by a number of acquisitions in different control groups at 6 and 14 months to assess the robustness of the procedure and its associated memory scores. These data illustrate the extraction of a variety of navigation scores (including search strategy, spatial learning and memory) and provide a reference of navigation data in the Starmaze task for healthy 6-month-old controls, normal aging and a model of pathological memory deficit.

• ApoE4 mice tested with spatial reversal learning present early flexibility deficit. • This flexibility deficit is not associated with memory impairment. • ApoE4 appears as a factor of early aging. To test the hypothesis that ApoE4 may be involved in cognitive deficits associated with aging, we investigated the impact of APOE4 status and aging on the flexibility and memory components of spatial learning in mice. Young adult (6 months) and middle-aged (14 months) ApoE4, ApoE3 and C57BL/6 male mice were tested for flexibility in an aquatic Y-maze, and for spatio-temporal memory acquisition in the Starmaze. Our results revealed a flexibility deficit of the 6-month-old ApoE4 mice compared to controls. However, this deficit was not associated with spatio-temporal memory deficit at the same age. Importantly, the ApoE4 flexibility deficit did not increase with age, nor turn into memory deficit, or was able to predict individual variations of memory performance at 14 months. By contrast, control ApoE3 mice showed a decline of flexibility at 14 months resulting in performance similar to that of ApoE4. Overall, our results suggest that ApoE4 could be associated with an acceleration of the flexibility decrease otherwise observed in normal aging. [ABSTRACT FROM AUTHOR]

Simulating human brain with hardware has been an attractive project for many years, since memory is one of the fundamental functions of our brains. Several memory models have been proposed up to now in order to unveil how the memory is organized in the brain. In this paper, we adopt spatio-temporal memory (STM) model, in which both associative memory and episodic memory are analyzed and emulated, as the reference of our hardware network architecture. Furthermore, some reasonable adaptations are carried out for the hardware implementation. We finally implement this memory model on FPGA, and additional experiments are performed to fine tune the parameters of our network deployed on FPGA.

Simulating human brain with hardware has been an attractive project for many years, since memory is one of the fundamental functions of our brains. Several memory models have been proposed up to now in order to unveil how the memory is organized in the brain. In this paper, we adopt spatio-temporal memory (STM) model, in which both associative memory and episodic memory are analyzed and emulated, as the reference of our hardware network architecture. Furthermore, some reasonable adaptations are carried out for the hardware implementation. We finally implement this memory model on FPGA, and additional experiments are performed to fine tune the parameters of our network deployed on FPGA. [ABSTRACT FROM AUTHOR]

Intelligent video anomaly detection (VAD) methods play a crucial role in conserving human resources, reducing the financial burden on governments, and promptly and accurately identifying abnormal behaviors. Frame prediction, which performs VAD by reasonably predicting normal video and distorting predicted anomalous video, is a popular and efficient method. Although Auto-Encoders (AE) show excellent performance in video frame prediction methods, the ability of these methods to use temporal information and poorly reconstruct anomalous videos is insufficient. To improve the shortcomings of AE in VAD, we propose a VAD method based on the cross-frame prediction mechanism and the spatio-temporal memory-enhanced pseudo-3D encoder. This method significantly enhances the recognition capability of abnormal activities in surveillance videos. Firstly, we use the prediction mechanism that uses multiple past frames with intervals to predict future frames. It can broaden the extra information and limit the memory consumption. Then we design the pseudo-3D encoder to encode the spatio-temporal information in videos, avoiding the problems that the 2D encoder cannot obtain the temporal dimensional information and the 3D encoder has complicated structure and overmuch parameters. Finally, we design the spatio-temporal memory block with three loss functions to store the spatio-temporal information of normal videos, which can expand the predicted differences between normal and abnormal examples. Experiments on UCSD Ped2, CUHK Avenue and ShanghaiTech datasets experimentally show that the proposed method achieves 99.4%, 90.5% and 74.3% of the AUC values. Our method shows excellent performance among single-stage semi-supervised anomaly detection methods. [ABSTRACT FROM AUTHOR]

Autonomous robots cohabiting with humans will have to achieve recurring tasks while adapting to the changing conditions of the world. A spatio-temporal memory categorizes the experiences of a robot to improve its ability to adapt to its environment. In this paper, we present a spatio-temporal (ST) memory model consisting of a cascade of two adaptive resonance theory (ART) networks: one to categorize spatial events and the other to extract temporal episodes from the robot's experiences. Artificial emotions are used to dynamically modulate learning and recall of the ART networks based on how the robot is able to carry its task, using a simple model of artificial emotions. Once an episode is recalled, future events can be predicted and used to influence the intentions of the robot. Evaluation of our ST model is done using an autonomous robotic platform that has to deliver objects to people within an office area. Results demonstrate that our model can memorize and recall the experiences of a robot, and that emotional episodes are recalled more often, allowing the robot to use its memory of past experiences early on when repeating a task. [ABSTRACT FROM AUTHOR]

Design of artificial neural structures capable of reliable and flexible long-term spatio-temporal memory is of paramount importance in machine intelligence. To this end, we propose a novel, biologically inspired, long-term memory (LTM) architecture. We intend to use it as a building block of a neuron-level architecture that is able to mimic natural intelligence through learning, anticipation, and goal-driven behavior. A mutual input enhancement and blocking structure is proposed, and its operation is discussed in detail. The paper focuses on a hierarchical memory organization, storage, recognition, and recall mechanisms. Simulation results of the proposed memory show its effectiveness, adaptability, and robustness. Accuracy of the proposed method is compared to other methods including Levenshtein distance method and a Markov chain. [ABSTRACT FROM AUTHOR]