A Cognitive Architecture Based on a Learning Classifier System with Spiking Classifiers

  title={A Cognitive Architecture Based on a Learning Classifier System with Spiking Classifiers},
  author={Gerard David Howard and Larry Bull and Pier Luca Lanzi},
  journal={Neural Processing Letters},
Learning classifier systems (LCS) are population-based reinforcement learners that were originally designed to model various cognitive phenomena. This paper presents an explicitly cognitive LCS by using spiking neural networks as classifiers, providing each classifier with a measure of temporal dynamism. We employ a constructivist model of growth of both neurons and synaptic connections, which permits a genetic algorithm to automatically evolve sufficiently-complex neural structures. The… 
Evolving Spiking Neural Networks for Nonlinear Control Problems
It is demonstrated that the proposed spiking controller can learn significantly faster to discover functional solutions than sigmoidal neural networks in solving a classic nonlinear control problem.
Accuracy-Based Learning Classifier Systems for Multistep Reinforcement Learning: A Fuzzy Logic Approach to Handling Continuous Inputs and Learning Continuous Actions
A new accuracy-based learning fuzzy classifier system (LFCS) is developed to explicitly handle continuous state input and continuous action output during multistep reinforcement learning and is shown to perform highly competitively on four benchmark learning problems and a robotics problem.
Deep Learning with a Classifier System: Initial Results
This article presents the first results from using a learning classifier system capable of performing adaptive computation with deep neural networks, and automatically reduces the number of weights and units while maintaining performance after achieving a maximum prediction error.
Autoencoding with XCSF
This article explores the use of the XCSF online evolutionary reinforcement learning system to perform autoencoding using a neural network representation and combining artificial evolution with stochastic gradient descent to suggest it is an effective approach to data reduction.
Autoencoding With a Classifier System
This article explores the use of an LCS to adaptively decompose the input domain into a collection of small autoencoders, where local solutions of different complexity may emerge, and shows possible to reduce the code size as well as the resulting decoder computational cost when compared with the global model equivalent.
Convergence analysis of rule-generality on the XCS classifier system
This analysis provides the first proof that an average rule-generality of rules in a population can converge to its optimum value under some assumptions and can be used to mathematically conclude that the XCS framework has a natural pressure to explore rules toward optimum rules if XCS satisfies the authors' derived conditions.
DeepExpert : vers une Intelligence Artificielle autonome et explicable
Pour procurer aux intelligences artificielles une meilleure adaptabilite, notre position est de proposer un couplage selon une approche enactiviste et cognitiviste entre reseaux de neurones profonds


A spiking neural representation for XCSF
A Learning Classifier System where each traditional rule is represented by a spiking neural network, a type of network with dynamic internal state, allowing for the evolution of networks of appropriate complexity to emerge whilst solving a continuous maze environment.
Evolution of Spiking Neural Controllers for Autonomous Vision-Based Robots
It is shown that evolution finds relatively quickly functional spiking controllers capable of navigating in irregularly textured environments without hitting obstacles using a very simple genetic encoding and fitness function.
An evolutionary cognitive architecture made of a bag of networks
A cognitive architecture is presented for modelling some properties of sensorimotor learning in infants, namely the ability to accumulate adaptations and skills over multiple tasks in a manner which
A Neural Learning Classifier System with Self-Adaptive Constructivism for Mobile Robot Control
This article examines the use of constructivism-inspired mechanisms within a neural learning classifier system architecture that exploits parameter self-adaptation as an approach to realize such behavior.
Controlling a Simulated Khepera with an XCS Classifier System with Memory
Results of using XCS with additional internal memory in the simulated Khepera environment are presented, and it is shown that control rules can be evolved to allow the robot to navigate a variety of problems.
A neural learning classifier system with self-adaptive constructivism
  • L. Bull, Jacob Hurst
  • Computer Science
    The 2003 Congress on Evolutionary Computation, 2003. CEC '03.
  • 2003
It is shown that appropriate internal rule complexity emerges during learning at a rate controlled by the learner and that the structure indicates underlying features of the task.
Towards continuous actions in continuous space and time using self-adaptive constructivism in neural XCSF
A Learning Classifier System (LCS) where each classifier condition is represented by a feed-forward multi-layered perceptron (MLP) network, allowing for the evolution of networks of appropriate complexity to solve a continuous maze environment.
Networks of Spiking Neurons: The Third Generation of Neural Network Models
  • W. Maass
  • Computer Science
    Electron. Colloquium Comput. Complex.
  • 1996