Learning to Act through Evolution of Neural Diversity in Random Neural Networks

Abstract

Biological nervous systems consist of networks of diverse, sophisticated information processors in the form of neurons of different classes. In most artificial neural networks (ANNs), neural computation is abstracted to an activation function that is usually shared between all neurons within a layer or even the whole network; training of ANNs focuses on synaptic optimization. In this paper, we propose the optimization of neuro-centric parameters to attain a set of diverse neurons that can perform complex computations. Demonstrating the promise of the approach, we show that evolving neural parameters alone allows agents to solve various reinforcement learning tasks without optimizing any synaptic weights. While not aiming to be an accurate biological model, parameterizing neurons to a larger degree than the current common practice, allows us to ask questions about the computational abilities afforded by neural diversity in random neural networks. The presented results open up interesting future research directions, such as combining evolved neural diversity with activity-dependent plasticity.

arXiv commentLinebreaks in abstract fixed

References in corpus (9)

• Deep Learning in Spiking Neural Networks
• The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
• Evolution Strategies as a Scalable Alternative to Reinforcement Learning
• OpenAI Gym
• Stabilizing the Lottery Ticket Hypothesis
• Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networks
• Evolving Normalization-Activation Layers
• Learning to learn with backpropagation of Hebbian plasticity
• Meta-Reinforcement Learning with Self-Modifying Networks

Cited by in corpus (1)

• Structurally Flexible Neural Networks: Evolving the Building Blocks for General Agents