Learning to Act through Evolution of Neural Diversity in Random Neural Networks
This work addresses the challenge of enhancing computational abilities in neural networks for AI researchers, though it is incremental as it builds on existing neural network paradigms.
The paper tackled the problem of limited neural diversity in artificial neural networks by proposing to optimize neuro-centric parameters to create diverse neurons, enabling agents to solve various reinforcement learning tasks without synaptic weight optimization.
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.