Implicit Regularization and Momentum Algorithms in Nonlinearly Parameterized Adaptive Control and Prediction

TL;DR

This paper explores how modern optimization techniques like natural gradient descent and mirror descent can be used to improve adaptive control and prediction in nonlinear dynamical systems, revealing implicit regularization effects.

Contribution

It introduces non-Euclidean adaptation laws inspired by optimization methods, demonstrating their regularization properties and developing a variational formalism with momentum for adaptive algorithms.

Findings

Implicit regularization selects desirable parameters among many solutions.

Non-Euclidean adaptation laws improve control and prediction accuracy.

Momentum-based laws derived from variational principles enhance convergence.

Abstract

Stable concurrent learning and control of dynamical systems is the subject of adaptive control. Despite being an established field with many practical applications and a rich theory, much of the development in adaptive control for nonlinear systems revolves around a few key algorithms. By exploiting strong connections between classical adaptive nonlinear control techniques and recent progress in optimization and machine learning, we show that there exists considerable untapped potential in algorithm development for both adaptive nonlinear control and adaptive dynamics prediction. We begin by introducing first-order adaptation laws inspired by natural gradient descent and mirror descent. We prove that when there are multiple dynamics consistent with the data, these non-Euclidean adaptation laws implicitly regularize the learned model. Local geometry imposed during learning thus may be used to select parameter vectors -- out of the many that will achieve perfect tracking or prediction -- for desired properties such as sparsity. We apply this result to regularized dynamics predictor and observer design, and as concrete examples, we consider Hamiltonian systems, Lagrangian systems, and recurrent neural networks. We subsequently develop a variational formalism based on the Bregman Lagrangian. We show that its Euler Lagrange equations lead to natural gradient and mirror descent-like adaptation laws with momentum, and we recover their first-order analogues in the infinite friction limit. We illustrate our analyses with simulations demonstrating our theoretical results.