Synaptic Activation and Dual Liquid Dynamics for Interpretable Bio-Inspired Models

TL;DR

This paper introduces a unified framework for bio-inspired RNN models that enhances interpretability and accuracy by incorporating liquid dynamics and chemical synapses, validated through a lane-keeping control task.

Contribution

It proposes a novel framework combining liquid dynamics and chemical synapses to improve interpretability and performance of bio-inspired RNNs.

Findings

Liquid-capacitance models yield interpretable dense RNNs.

Chemical synapses enhance interpretability.

Combined models achieve the best accuracy and interpretability.

Abstract

In this paper, we present a unified framework for various bio-inspired models to better understand their structural and functional differences. We show that liquid-capacitance-extended models lead to interpretable behavior even in dense, all-to-all recurrent neural network (RNN) policies. We further demonstrate that incorporating chemical synapses improves interpretability and that combining chemical synapses with synaptic activation yields the most accurate and interpretable RNN models. To assess the accuracy and interpretability of these RNN policies, we consider the challenging lane-keeping control task and evaluate performance across multiple metrics, including turn-weighted validation loss, neural activity during driving, absolute correlation between neural activity and road trajectory, saliency maps of the networks' attention, and the robustness of their saliency maps measured by the structural similarity index.