Evolving to learn: discovering interpretable plasticity rules for spiking networks

Abstract

Continuous adaptation allows survival in an ever-changing world. Adjustments in the synaptic coupling strength between neurons are essential for this capability, setting us apart from simpler, hard-wired organisms. How these changes can be mathematically described at the phenomenological level, as so called "plasticity rules", is essential both for understanding biological information processing and for developing cognitively performant artificial systems. We suggest an automated approach for discovering biophysically plausible plasticity rules based on the definition of task families, associated performance measures and biophysical constraints. By evolving compact symbolic expressions we ensure the discovered plasticity rules are amenable to intuitive understanding, fundamental for successful communication and human-guided generalization. We successfully apply our approach to typical learning scenarios and discover previously unknown mechanisms for learning efficiently from rewards, recover efficient gradient-descent methods for learning from target signals, and uncover various functionally equivalent STDP-like rules with tuned homeostatic mechanisms.

arXiv comment33 pages, 10 figures; J. Jordan and M. Schmidt contributed equally to this work

References in corpus (17)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• The NumPy array: a structure for efficient numerical computation
• A scalable multi-core architecture with heterogeneous memory structures for Dynamic Neuromorphic Asynchronous Processors (DYNAPs)
• The Evolved Transformer
• The Optimal Reward Baseline for Gradient-Based Reinforcement Learning
• Network Plasticity as Bayesian Inference
• AutoML-Zero: Evolving Machine Learning Algorithms From Scratch
• Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networks
• SpiNNaker 2: A 10 Million Core Processor System for Brain Simulation and Machine Learning
• Neuromorphic Hardware learns to learn
• Versatile emulation of spiking neural networks on an accelerated neuromorphic substrate
• STDP allows close-to-optimal spatiotemporal spike pattern detection by single coincidence detector neurons
• Stochasticity from function -- why the Bayesian brain may need no noise
• Transient chaotic dimensionality expansion by recurrent networks
• NESTML: a modeling language for spiking neurons
• Deterministic networks for probabilistic computing
• Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselves