Mistake gating leads to energy and memory efficient continual learning
Preprint
Abstract
Synaptic plasticity is metabolically expensive, yet animals continuously update their internal models without exhausting energy reserves. However, when artificial neural networks are trained, the network parameters are typically updated on every sample that is presented, even if the sample was classified correctly. Inspired by the human negativity bias and error-related negativity, we propose ‘memorized mistake-gated learning’—a biologically plausible plasticity rule where synaptic updates are strictly gated by current and past classification errors. This reduces the number of updates the network needs to make by 50% ∼ 80%. Mistake gating is particularly well suited in two cases: 1) For incremental learning where new knowledge is acquired on a background of pre-existing knowledge, 2) For online learning scenarios when data needs to be stored for later replay, as mistake-gating reduces storage buffer requirements. The algorithm can be implemented in a few lines of code, adds no hyper-parameters, and comes at negligible computational overhead. Learning on mistakes is an energy efficient and biologically relevant modification to commonly used learning rules that is well suited for continual learning.Keywords
All Related Versions
This publication has 23 references indexed in Scilit:
- Mitochondrial Ca2+ efflux controls neuronal metabolism and long-term memory across speciesNature Metabolism, 2026
- Selective Freezing for Efficient Continual Learning2023
- Energetically efficient learning in neuronal networksCurrent Opinion in Neurobiology, 2023
- Prediction errors disrupt hippocampal representations and update episodic memoriesProceedings of the National Academy of Sciences, 2021
- Frontal theta band oscillations predict error correction and posterror slowing in typing.Journal of Experimental Psychology: Human Perception and Performance, 2018
- Deep learningNature, 2015
- Classification-based objective functionsMachine Learning, 2006
- Heterosynaptic co-activation of glutamatergic and dopaminergic afferents is required to induce persistent long-term potentiationNeuropharmacology, 2004
- The neural basis of human error processing: Reinforcement learning, dopamine, and the error-related negativity.Psychological Review, 2002
- The perceptron: A probabilistic model for information storage and organization in the brain.Psychological Review, 1958