Memory Augmented Neural Network Adaptive Controllers: Performance and Stability

TL;DR

This paper introduces a neuroscience-inspired neural network adaptive control architecture with external memory, demonstrating improved estimation accuracy, stability, and learning performance through theoretical analysis and extensive simulations.

Contribution

It proposes a novel memory-augmented neural network control architecture, providing theoretical stability proofs and empirical evidence of enhanced learning and performance.

Findings

Memory augmentation improves estimation accuracy.

The controller achieves Uniformly Bounded stability.

Enhanced learning performance shown through simulations.

Abstract

In this paper, we propose a novel control architecture, inspired from neuroscience, for adaptive control of continuous-time systems. The proposed architecture, in the setting of standard Neural Network (NN) based adaptive control, augments an external working memory to the NN. The controller, through a write operation, writes the hidden layer feature vector of the NN to the external working memory and can also update this information with the observed error in the output. Through a read operation, the controller retrieves information from the working memory to modify the final control signal. First, we consider a simpler estimation problem to theoretically study the effect of an external memory and prove that the estimation accuracy can be improved by incorporating memory. We then consider a model reference NN adaptive controller for linear systems with matched uncertainty to implement and illustrate our ideas. We prove that the resulting controller leads to a Uniformly Bounded (UB) stable closed loop system. Through extensive simulations and specific metrics, such as peak deviation and settling time, we show that memory augmentation improves learning significantly. Importantly, we also provide evidence for and insights on the mechanism by which this specific memory augmentation improves learning.