Current State and Future Directions for Learning in Biological Recurrent Neural Networks: A Perspective Piece

Abstract

This perspective piece came about through the Generative Adversarial Collaboration (GAC) series of workshops organized by the Computational Cognitive Neuroscience (CCN) conference in 2020. We brought together a number of experts from the field of theoretical neuroscience to debate emerging issues in our understanding of how learning is implemented in biological recurrent neural networks. Here, we will give a brief review of the common assumptions about biological learning and the corresponding findings from experimental neuroscience and contrast them with the efficiency of gradient-based learning in recurrent neural networks commonly used in artificial intelligence. We will then outline the key issues discussed in the workshop: synaptic plasticity, neural circuits, theory-experiment divide, and objective functions. Finally, we conclude with recommendations for both theoretical and experimental neuroscientists when designing new studies that could help to bring clarity to these issues.

Probing learning through the lens of changes in circuit dynamics

“…Despite exciting experimental efforts in building causal approaches of linking synaptic plasticity to behavioral measures of learning, bridging the gap between the two remains a challenge (30,31). In parallel, theories for learning algorithms are mushrooming (32)(33)(34), with little experimental evidence to disambiguate between them. We desperately need new ideas to approach the question of learning and its neural correlates, in ways that are more experimentally tractable.…”