Evolving and Merging Hebbian Learning Rules: Increasing Generalization by Decreasing the Number of Rules
This work addresses generalization and overfitting issues in artificial agents, particularly for robotics applications, by introducing a regularizing 'genomic bottleneck' approach, though it is incremental as it builds on existing Hebbian learning methods.
The paper tackled the challenge of generalization to out-of-distribution (OOD) circumstances in plastic Hebbian neural networks by evolving and merging Hebbian learning rules to reduce trainable parameters, resulting in a reduction from 61,440 to 1,920 parameters while improving robustness on 30 unseen robot morphologies.
Generalization to out-of-distribution (OOD) circumstances after training remains a challenge for artificial agents. To improve the robustness displayed by plastic Hebbian neural networks, we evolve a set of Hebbian learning rules, where multiple connections are assigned to a single rule. Inspired by the biological phenomenon of the genomic bottleneck, we show that by allowing multiple connections in the network to share the same local learning rule, it is possible to drastically reduce the number of trainable parameters, while obtaining a more robust agent. During evolution, by iteratively using simple K-Means clustering to combine rules, our Evolve and Merge approach is able to reduce the number of trainable parameters from 61,440 to 1,920, while at the same time improving robustness, all without increasing the number of generations used. While optimization of the agents is done on a standard quadruped robot morphology, we evaluate the agents' performances on slight morphology modifications in a total of 30 unseen morphologies. Our results add to the discussion on generalization, overfitting and OOD adaptation. To create agents that can adapt to a wider array of unexpected situations, Hebbian learning combined with a regularising "genomic bottleneck" could be a promising research direction.