Learning Linear Dynamical Systems with Semi-Parametric Least Squares

Abstract

We analyze a simple prefiltered variation of the least squares estimator for the problem of estimation with biased, semi-parametric noise, an error model studied more broadly in causal statistics and active learning. We prove an oracle inequality which demonstrates that this procedure provably mitigates the variance introduced by long-term dependencies. We then demonstrate that prefiltered least squares yields, to our knowledge, the first algorithm that provably estimates the parameters of partially-observed linear systems that attains rates which do not not incur a worst-case dependence on the rate at which these dependencies decay. The algorithm is provably consistent even for systems which satisfy the weaker marginal stability condition obeyed by many classical models based on Newtonian mechanics. In this context, our semi-parametric framework yields guarantees for both stochastic and worst-case noise.

References in corpus (6)

• The Optimal Reward Baseline for Gradient-Based Reinforcement Learning
• REBAR: Low-variance, unbiased gradient estimates for discrete latent variable models
• Gradient Descent Learns Linear Dynamical Systems
• Online Least Squares Estimation with Self-Normalized Processes: An Application to Bandit Problems
• Variance Reduction Methods for Sublinear Reinforcement Learning
• Semiparametric Contextual Bandits

Cited by in corpus (17)

• Logarithmic Regret Bound in Partially Observable Linear Dynamical Systems
• Active Learning for Nonlinear System Identification with Guarantees
• The Nonstochastic Control Problem
• Making Non-Stochastic Control (Almost) as Easy as Stochastic
• Efficient Learning of Distributed Linear-Quadratic Controllers
• Fairness in Forecasting and Learning Linear Dynamical Systems
• Sample Complexity of Kalman Filtering for Unknown Systems
• No-Regret Prediction in Marginally Stable Systems
• Fairness in Forecasting of Observations of Linear Dynamical Systems
• PAC-Bayes Generalisation Bounds for Dynamical Systems Including Stable RNNs
• Non-Stochastic Control with Bandit Feedback
• Certainty Equivalent Perception-Based Control
• Robust guarantees for learning an autoregressive filter
• Task-Optimal Exploration in Linear Dynamical Systems
• Learning Partially Observed Linear Dynamical Systems from Logarithmic Number of Samples
• Improved rates for prediction and identification of partially observed linear dynamical systems
• Statistical Consistency of Set-Membership Estimator for Linear Systems