{"title":"Pattern Formation in CNN Working on the Edge of Chaos","authors":"A. Slavova, Ventsislav Ignatov","doi":"10.1109/CNNA49188.2021.9610771","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610771","url":null,"abstract":"In this paper we present a special class of Cellular Nonlinear/Nanoscale Networks (CNN) operating on the edge of chaos. We study the dynamics of the model for pattern formation via local activity theory. We determine the edge of chaos region in which complex behavior of our model arises. As an application we consider image coding method which is based on pattern generation of CNN structures under consideration.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122429134","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":"On The Resilience of Cellular Neural Networks to Low-intensity Adversarial Attacks","authors":"A. Horváth","doi":"10.1109/CNNA49188.2021.9610769","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610769","url":null,"abstract":"Deep Neural networks are commonly used in various tasks and enabled the solution of many practical problems. These approaches usually result sufficiently high accuracy, but the robustness of these methods in critical applications is still under investigation. Adversarial attacks, in which minor perturbations can cause misclassification pose one of the most significant challenges. In case of convolutional neural networks there is ongoing research to create more resilient networks towards these attacks. In this paper I will demonstrate that multi-layered cellular neural networks in their nature are more robust and resilient to low-intensity attacks than their convolutional counterparts.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121270068","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":"A novel hardware-efficient asynchronous cellular automaton model of tumor immunotherapy and its FPGA implementation","authors":"Naoto Horie, H. Torikai","doi":"10.1109/CNNA49188.2021.9610753","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610753","url":null,"abstract":"A novel tumor immunotherapy model based on the nonlinear dynamics of an asynchronous cellular automaton is presented. It is shown that the presented model can reproduce basic bifurcation structure of a differential equation tumor immunotherapy model qualitatively. In addition, the presented model is implemented on a field programmable gate array and experiments validate occurrence of the basic bifurcation. It is also shown that the presented model can be implemented using fewer hardware resources and lower power than the differential equation model.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116104715","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":"Sand Castle Summation For Pixel Processor Arrays","authors":"Laurie Bose, P. Dudek, Jianing Chen, S. Carey","doi":"10.1109/CNNA49188.2021.9610764","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610764","url":null,"abstract":"Pixel Processor Arrays (PPA) present a new vision sensor/processor architecture consisting of a SIMD array of processor elements, each capable of light capture, storage, processing and local communication. Such a device allows visual data to be efficiently stored and manipulated directly upon the focal plane, but also demands the invention of new approaches and algorithms, suitable for the massively-parallel fine-grain processor arrays. In this paper we implement an image-wide population count algorithm exploiting the parallel processing of the PPA. Performing such a global count was previously unviable for vision processing tasks due to its exhaustive computation time. Our approach shows an improvement of typically two orders of magnitude reduction in computation time, thus allowing it to be incorporated as a core component of many vision tasks upon the PPA.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123801968","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":"Mathematical Investigation of Static Pattern Formation with a Locally Active Memristor Model","authors":"A. S. Demirkol, A. Ascoli, R. Tetzlaff","doi":"10.1109/CNNA49188.2021.9610811","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610811","url":null,"abstract":"We present the mathematical investigation of static pattern formation in a Memristor Cellular Nonlinear Network (M -CNN), in consideration of the theory of local activity. The M-CNN has a planar grid form composed of identical memristive cells, which are purely resistively coupled to each other. The single cell contains a DC voltage source, a bias resistor, and a locally active memristor in parallel with a capacitor. The memristor model employed has a simple generic form which helps to reduce the simulation time, and has a functional AC equivalent circuit which facilitates further calculations. We adopt a circuit theoretical approach for the stability analysis of the single cell and a 3-cell ring configuration, as well as the examination of local activity, edge-of-chaos, and sharp-edge-of-chaos domains, which helps us to interpret the results in a better way. The emergence of static patterns is successfully confirmed by simulating the proposed resistively coupled M -CNN utilizing locally active memristors.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"176 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124330570","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":"On controlling multistability in memristor circuits","authors":"M. Di Marco, M. Forti, G. Innocenti, A. Tesi","doi":"10.1109/CNNA49188.2021.9610751","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610751","url":null,"abstract":"One of the appealing properties of memristor circuits is multistability, i.e., the coexistence in the state space of a large variety of different attractors. Since each attractor belongs to one of the infinitely many invariant manifolds in which the state space is decomposed, switching between different attractors requires to address the problem of steering the memristor circuits dynamics from one invariant manifold to another in a given time interval. For a fairly general class of circuits containing an ideal memristor it is shown that it is always possible to solve this problem via a pulse programmed source.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115878778","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":"Analysis of time series classification of a multi-layer reservoir neural network based on asynchronous cellular automaton neurons with transmission delays","authors":"Kohei Nakata, H. Torikai","doi":"10.1109/CNNA49188.2021.9610744","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610744","url":null,"abstract":"In this paper, a novel multi-layer reservoir neural network with axonal delays is proposed using an asynchronous cellular automaton neuron model. A learning method of the network based on the simulated annealing is also proposed. Then, performance of time series classification of the network is analyzed with respect to parameters of the reservoir layers. Based on the analysis results, a design method of the network to realize higher performance of the time series classification is proposed. Furthermore, the proposed network is implemented as a hardware description language code (Verilog-HDL code) and post-synthesize simulations validate its classification function.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116016225","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":"Heteroclinic cycles in Chua-Yang ring networks","authors":"M. Koller, Marcell Simkó, B. Garay","doi":"10.1109/CNNA49188.2021.9610774","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610774","url":null,"abstract":"Heteroclinic equilibrium connections in Chua-Yang ring networks with two-parameter nearest-neighbor coupling are investigated. The dynamics is governed by the standard piecewise linear saturated activation function. We look for heteroclinic cycles created by terminating families of Hopf periodic orbits. Our approach is based on numerical experimentation supported by rigorous a posteriori mathematical arguments.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128460959","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":"Comparing Different PC and FPGA Implementation Possibilities of Fast Multipole Method","authors":"Levente Santha, Z. Nagy, A. Kiss, G. Csaba","doi":"10.1109/CNNA49188.2021.9610776","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610776","url":null,"abstract":"In this paper our aim is to compare the efficiency of different hardware platforms during the solution of the Fast Multipole Method (FMM). The brute force solution has also been compared to the highly topographical FMM algorithm to reflect the advantages and disadvantages of the concept. During the implementation we attempted to benefit from the different properties of the hardware platforms (e.g.: multi computation cores in PCs and array clusters in Field Programmable Gate Arrays - FPGAs). We demonstrate different implementations on PC and on FPGA with high-level hardware synthesis and benchmark the resulting hardware in terms of speed, and power consumption.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126729883","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":"Breaking the Sensorimotor Loop - A Memristor-Ready Robot Control Architecture","authors":"M. Hild, Maximilian Tolksdorf, Benjamin Panreck","doi":"10.1109/CNNA49188.2021.9610795","DOIUrl":"https://doi.org/10.1109/CNNA49188.2021.9610795","url":null,"abstract":"Traditional sensorimotor loops are causal, i.e., the underlying control algorithm exhibits separate input and output lines, where the output fully depends on the input and an internal state of the algorithm itself. This is both true for digital and analog implementations. In the paper at hand we propose an acausal cellular analog architecture for robot control which offers advantages when memristors are to be incorporated for behavior switching and adaptation. Even for complex behavior, e.g., a robot standing-up, the architecture can stay simple since no cross-connections between the different joints' motor control units are needed.","PeriodicalId":325231,"journal":{"name":"2021 17th International Workshop on Cellular Nanoscale Networks and their Applications (CNNA)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115472457","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}