Timezone: »

 
Poster
On the Convergence of SARSA with Linear Function Approximation
Shangtong Zhang · Remi Tachet des Combes · Romain Laroche

Wed Jul 26 05:00 PM -- 06:30 PM (PDT) @ Exhibit Hall 1 #222
in Poster Session 4 »

SARSA, a classical on-policy control algorithm for reinforcement learning, is known to chatter when combined with linear function approximation: SARSA does not diverge but oscillates in a bounded region. However, little is known about how fast SARSA converges to that region and how large the region is. In this paper, we make progress towards this open problem by showing the convergence rate of projected SARSA to a bounded region. Importantly, the region is much smaller than the region that we project into, provided that the the magnitude of the reward is not too large. Existing works regarding the convergence of linear SARSA to a fixed point all require the Lipschitz constant of SARSA's policy improvement operator to be sufficiently small; our analysis instead applies to arbitrary Lipschitz constants and thus characterizes the behavior of linear SARSA for a new regime.

[ PDF]  [ Poster] 

Author Information

Shangtong Zhang (University of Virginia)
Remi Tachet des Combes (AlpacaML)
Romain Laroche (None)

More from the Same Authors