Sample complexity and regret bounds are given that depend on the so-called eluder dimension of the function set and are independent of other parameters. The agnostic case is considered as well.

In this paper we complement these previous studies of operator alternatives and focus on operators that seek to achieve greater robustness w.r.t.

Neural Information Processing Systems http://nips.cc/

Due to state transitions, it is challenging to balance the contribution from each dimension for achieving near-optimality.

SARSA is an on-policy algorithm to learn a Markov decision process policy in reinforcement learning. We investigate the SARSA algorithm with linear function approximation under the non-i.i.d.

A unique benefit to this approach is that each transition's TD error can be