Timezone: »

 
Spotlight
Regret Bounds for Stochastic Shortest Path Problems with Linear Function Approximation
Daniel Vial · Advait Parulekar · Sanjay Shakkottai · R Srikant

Wed Jul 20 01:50 PM -- 01:55 PM (PDT) @ Ballroom 3 & 4
in Theory »

We propose an algorithm that uses linear function approximation (LFA) for stochastic shortest path (SSP). Under minimal assumptions, it obtains sublinear regret, is computationally efficient, and uses stationary policies. To our knowledge, this is the first such algorithm in the LFA literature (for SSP or other formulations). Our algorithm is a special case of a more general one, which achieves regret square root in the number of episodes given access to a computation oracle.

Keywords: T: Reinforcement Learning and Planning

Author Information

Daniel Vial (UT Austin / UIUC)
Advait Parulekar (University of Texas at Austin)
Sanjay Shakkottai (University of Texas at Austin)
R Srikant (UIUC)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors