Neuroevolution Of Recurrent Architectures On Control Tasks

Abstract

Modern artificial intelligence works typically train the parameters of fixed-sized deep neural networks using gradient-based optimization techniques. Simple evolutionary algorithms have recently been shown to also be capable of optimizing deep neural network parameters, at times matching the performance of gradient-based techniques, e.g. in reinforcement learning settings. In addition to optimizing network parameters, many evolutionary computation techniques are also capable of progressively constructing network architectures. However, constructing network architectures from elementary evolution rules has not yet been shown to scale to modern reinforcement learning benchmarks. In this paper we therefore propose a new approach in which the architectures of recurrent neural networks dynamically evolve according to a small set of mutation rules. We implement a massively parallel evolutionary algorithm and run experiments on all 19 OpenAI Gym state-based reinforcement learning control task

Stats

• citations3
• S2 citations—
• github stars0
• HF likes0
• heat score4.52
• arxiv keyclei2023neuroevolution

Related papers

• Neural Architecture Evolution In Deep Reinforcement Learning For Continuous Control (2019)0.00
• Efficacy Of Modern Neuro-evolutionary Strategies For Continuous Control Optimization (2019)0.00
• Learning To Act Through Evolution Of Neural Diversity In Random Neural Networks (2023)2.26
• Evolving-to-learn Reinforcement Learning Tasks With Spiking Neural Networks (2022)0.00
• Reducing Catastrophic Forgetting When Evolving Neural Networks (2019)0.00
• Mimicking Evolution With Reinforcement Learning (2020)0.00
• Optimizing The Neural Architecture Of Reinforcement Learning Agents (2020)0.00
• Collaborative Evolutionary Reinforcement Learning (2019)0.00