Timezone: »

 
Big Control Actions Help Multitask Learning of Unstable Linear Systems
Aditya Modi · Ziping Xu · Mohamad Kazem Shirani Faradonbeh · Ambuj Tewari

@
in Complex feedback in online learning »

A fundamental problem in one of the most popular models for continuous environments; i.e., linear dynamical systems, is to learn dynamics matrices from unstable state trajectories. While this problem is well-studied for a single system, little is known about multitask learning methods that utilize potential commonalities to estimate the dynamics matrices more accurately. The longstanding obstacle is that idiosyncratic instabilities nullify the benefit of sharing data across systems. We address this issue by introducing the new method of \emph{big random control actions}, and develop a novel performance analysis for that. To the authors' knowledge, this is the first theoretical guarantee for multitask learning under instability when system matrices are \emph{unknown} linear combinations of \emph{unknown} bases matrices. The techniques can be extended to multitask learning problems in other settings with non-stationary or temporally dependent data.

Author Information

Aditya Modi (Microsoft)
Ziping Xu (University of Michigan)
Mohamad Kazem Shirani Faradonbeh (University of Georgia)
Ambuj Tewari (University of Michigan)

More from the Same Authors