A Spiking Neural Learning Classifier System

2012/01/16 by Howard, Gerard, Larry Bull, Pier Luca Lanzi +2
Computer Science · Engineering · #Advanced Memory and Neural Computing #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Neural Networks and Applications #Neural and Evolutionary Computing (cs.NE) #Robotics (cs.RO)

paper · pdf · doi:10.48550/arxiv.1201.3249

openalex publication_date 2012/01/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Learning Classifier Systems (LCS) are population-based reinforcement learners used in a wide variety of applications. This paper presents a LCS where each traditional rule is represented by a spiking neural network, a type of network with dynamic internal state. We employ a constructivist model of growth of both neurons and dendrites that realise flexible learning by evolving structures of sufficient complexity to solve a well-known problem involving continuous, real-valued inputs. Additionally, we extend the system to enable temporal state decomposition. By allowing our LCS to chain together sequences of heterogeneous actions into macro-actions, it is shown to perform optimally in a problem where traditional methods can fail to find a solution in a reasonable amount of time. Our final system is tested on a simulated robotics platform.

Citations

• Wrappers for feature subset selection
• Robot shaping: developing autonomous agents through learning
• Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning

Related

• A Cognitive Architecture Based on a Learning Classifier System with Spiking Classifiers
• A Spiking Neural Network Learning Markov Chain
• A Spiking Neural Network Structure Implementing Reinforcement Learning
• Fully Spiking Actor Network with Intra-layer Connections for Reinforcement Learning
• Population-coding and Dynamic-neurons improved Spiking Actor Network for Reinforcement Learning