Predictive Learning in Energy-based Models with Attractor Structures

TL;DR

This paper presents a biologically plausible energy-based neural model that predicts observations after actions, integrating hierarchical structures and attractor networks, and demonstrating effectiveness across diverse scenarios.

Contribution

It introduces a novel energy-based model with attractor structures for neural prediction, bridging biological plausibility and machine learning performance.

Findings

Accurately predicts environment changes in trained scenarios

Generalizes well to unseen environments

Matches machine learning methods in diverse tasks

Abstract

Predictive models are highly advanced in understanding the mechanisms of brain function. Recent advances in machine learning further underscore the power of prediction for optimal representation in learning. However, there remains a gap in creating a biologically plausible model that explains how the neural system achieves prediction. In this paper, we introduce a framework that employs an energy-based model (EBM) to capture the nuanced processes of predicting observation after action within the neural system, encompassing prediction, learning, and inference. We implement the EBM with a hierarchical structure and integrate a continuous attractor neural network for memory, constructing a biologically plausible model. In experimental evaluations, our model demonstrates efficacy across diverse scenarios. The range of actions includes eye movement, motion in environments, head turning, and static observation while the environment changes. Our model not only makes accurate predictions for environments it was trained on, but also provides reasonable predictions for unseen environments, matching the performances of machine learning methods in multiple tasks. We hope that this study contributes to a deep understanding of how the neural system performs prediction.