Chaosnet: a chaos based artificial neural network architecture for classification

DOI10.1063/1.5120831zbMATH Open1427.92005arXiv1910.02423OpenAlexW2989628193WikidataQ91515366 ScholiaQ91515366MaRDI QIDQ5205680FDOQ5205680

Authors: Harikrishnan Nellippallil Balakrishnan, Aditi Kathpalia, Snehanshu Saha, Nithin Nagaraj 

Publication date: 13 December 2019

Published in: Chaos: An Interdisciplinary Journal of Nonlinear Science (Search for Journal in Brave)

Abstract: Inspired by chaotic firing of neurons in the brain, we propose ChaosNet -- a novel chaos based artificial neural network architecture for classification tasks. ChaosNet is built using layers of neurons, each of which is a 1D chaotic map known as the Generalized Luroth Series (GLS) which has been shown in earlier works to possess very useful properties for compression, cryptography and for computing XOR and other logical operations. In this work, we design a novel learning algorithm on ChaosNet that exploits the topological transitivity property of the chaotic GLS neurons. The proposed learning algorithm gives consistently good performance accuracy in a number of classification tasks on well known publicly available datasets with very limited training samples. Even with as low as 7 (or fewer) training samples/class (which accounts for less than 0.05% of the total available data), ChaosNet yields performance accuracies in the range 73.89 % - 98.33 %. We demonstrate the robustness of ChaosNet to additive parameter noise and also provide an example implementation of a 2-layer ChaosNet for enhancing classification accuracy. We envisage the development of several other novel learning algorithms on ChaosNet in the near future.

Full work available at URL: https://arxiv.org/abs/1910.02423

Recommendations

• scientific article; zbMATH DE number 14852
• Annealed chaotic neural network with nonlinear self-feedback and its application to clustering problem
• Learning-induced pattern classification in a chaotic neural network
• Robust chaos in neural networks
• Neural networks and chaos: construction, evaluation of chaotic networks, and prediction of chaos with multilayer feedforward networks

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Strange attractors, chaotic dynamics of systems with hyperbolic behavior (37D45) Neural networks for/in biological studies, artificial life and related topics (92B20)

Cites Work

• Scikit-learn: machine learning in Python
• Nearest neighbor pattern classification
• Title not available (Why is that?)
• Approximation by superpositions of a sigmoidal function
• Clustering, coding, switching, hierarchical ordering, and control in a network of chaotic elements
• Chaos
• Bifurcations, burstings, chaos and crises in the Rose-Hindmarsh model for neuronal activity
• A chaotic model of sustaining attention problem in attention deficit disorder
• Chaotic dynamics of information processing: The magic number sevel plus- minus two revisited
• Using chaotic artificial neural networks to model memory in the brain
• On the information bottleneck theory of deep learning

Cited In (7)

• Title not available (Why is that?)
• New 4D and 3D models of chaotic systems developed from the dynamic behavior of nuclear reactors
• Title not available (Why is that?)
• When noise meets chaos: stochastic resonance in neurochaos learning
• Neural networks and chaos: construction, evaluation of chaotic networks, and prediction of chaos with multilayer feedforward networks
• Multistable dynamics and control of a new 4D memristive chaotic Sprott B system
• Stimulus classification using chimera-like states in a spiking neural network

Uses Software

• UCI-ml
• Scikit
• MNIST
• Keras

This page was built for publication: Chaosnet: a chaos based artificial neural network architecture for classification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5205680)