{"title":"Dissipative State Estimation for Delayed Discrete-Time Memristive Neural Networks Under Dynamic Event-Triggered Mechanisms and Deception Attacks","authors":"Huimin Tao, Hongjian Liu, Jiali Zhu, Hailong Tan","doi":"10.1109/icnc52316.2021.9607893","DOIUrl":"https://doi.org/10.1109/icnc52316.2021.9607893","url":null,"abstract":"In this paper, the state estimation problem for delayed discrete-time memristive neural networks subject to deception attacks is concerned. The main purpose is to design a state estimator such that the error dynamics of the state estimation is stochastically stable and strict dissipative. The randomness of deception attacks is described via a set of Bernoulli distributed white sequences with known conditional probabilities. In addition, in order to reduce unnecessary resource consumption in the communication channels, a dynamic event-triggering mechanism is introduced. By using the Lyapunov-Krasovskii functional and stochastic analysis method, sufficient conditions for stochastic stability and strict dissipation of the error system are acquired. On this basis, a co-design scheme of estimator parameters and the event-triggered matrix is proposed. Finally, the numerical simulation shows the effectiveness of the proposed state estimation scheme.","PeriodicalId":412866,"journal":{"name":"2021 International Conference on Neuromorphic Computing (ICNC)","volume":"188 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122182325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Predefined-Time Synchronization of Inertial Memristive Neural Networks","authors":"Kan Zeng, Leimin Wang","doi":"10.1109/icnc52316.2021.9608785","DOIUrl":"https://doi.org/10.1109/icnc52316.2021.9608785","url":null,"abstract":"This article investigates the predefined-time synchronization (PTS) of inertial memristive neural networks (IMNNs). First, we reduce the order of the system and design an effective controller for the error system. Furthermore, several new criteria for PTS of IMNNs are obtained and a user-defined settling time is provided. Lastly, the effectiveness of the results is proved by simulation experiment.","PeriodicalId":412866,"journal":{"name":"2021 International Conference on Neuromorphic Computing (ICNC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123241705","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Bipolar Pulse-induced Coexisting Symmetric Attractors in Hopfield Neural Network","authors":"Chengjie Chen, Fuhong Min","doi":"10.1109/icnc52316.2021.9608135","DOIUrl":"https://doi.org/10.1109/icnc52316.2021.9608135","url":null,"abstract":"By exerting a bipolar pulse (BP) current on one neuron, a non-autonomous Hopfield neural network (HNN) model is proposed in this paper, where complex dynamical effects of coexisting symmetric attractors are exhibited apparently. Firstly, using graphical method to calculate the switchable equilibrium points, we find that unstable node-focus (UNF) and stable node-focus (SNF) are existed in the model. Additionally, using multiple dynamical methods to reveal the initial-related bistable patterns, we find that the proposed model behaves coexisting symmetric attractors. Finally, we develop an electronic neuron circuit for the model on the analog level, and the PSIM-based circuit simulations excellently verify the numerical results.","PeriodicalId":412866,"journal":{"name":"2021 International Conference on Neuromorphic Computing (ICNC)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121535241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Memristive Network-based Artificial Fish Swarm Algorithm","authors":"Mingxiao Wang, Yongbin Yu, Quan-xin Deng, Chen Zhou","doi":"10.1109/icnc52316.2021.9607905","DOIUrl":"https://doi.org/10.1109/icnc52316.2021.9607905","url":null,"abstract":"This paper explores the relation between Artificial Fish Swarm Algorithm and memristive network. There are some relations between swarm intelligence and memristive network. Inspired by the correlation between colony optimization and memristive network, mapping rules between AFSA's 4 main behaviors and memristive network has highlighted in this paper. In this paper the motion of single fish is first simulated, then the motion of more fish is simulated. Finally, we make comparison of the trend of motion fish swarm and concept model.","PeriodicalId":412866,"journal":{"name":"2021 International Conference on Neuromorphic Computing (ICNC)","volume":"256 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121882860","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Victor Caceres Chian, Marcel Hildebrandt, T. Runkler, D. Dold
{"title":"Learning Through Structure: Towards Deep Neuromorphic Knowledge Graph Embeddings","authors":"Victor Caceres Chian, Marcel Hildebrandt, T. Runkler, D. Dold","doi":"10.1109/ICNC52316.2021.9607968","DOIUrl":"https://doi.org/10.1109/ICNC52316.2021.9607968","url":null,"abstract":"Computing latent representations for graph-structured data is an ubiquitous learning task in many industrial and academic applications ranging from molecule synthetization to social network analysis and recommender systems. Knowledge graphs are among the most popular and widely used data representations related to the Semantic Web. Next to structuring factual knowledge in a machine-readable format, knowledge graphs serve as the backbone of many artificial intelligence applications and allow the ingestion of context information into various learning algorithms. Graph neural networks attempt to encode graph structures in low-dimensional vector spaces via a message passing heuristic between neighboring nodes. Over the recent years, a multitude of different graph neural network architectures demonstrated ground-breaking performances in many learning tasks. In this work, we propose a strategy to map deep graph learning architectures for knowledge graph reasoning to neuromorphic architectures. Based on the insight that randomly initialized and untrained (i.e., frozen) graph neural networks are able to preserve local graph structures, we compose a frozen neural network with shallow knowledge graph embedding models. We experimentally show that already on conventional computing hardware, this leads to a significant speedup and memory reduction while maintaining a competitive performance level. Moreover, we extend the frozen architecture to spiking neural networks, introducing a novel, event-based and highly sparse knowledge graph embedding algorithm that is suitable for implementation in neuromorphic hardware.","PeriodicalId":412866,"journal":{"name":"2021 International Conference on Neuromorphic Computing (ICNC)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125104213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}