Improving reinforcement learning algorithms: towards optimal learning rate policies (1911.02319v6)

Abstract: This paper investigates to what extent one can improve reinforcement learning algorithms. Our study is split in three parts. First, our analysis shows that the classical asymptotic convergence rate $O(1/\sqrt{N})$ is pessimistic and can be replaced by $O((\log(N)/N)^{\beta})$ with $\frac{1}{2}\leq \beta \leq 1$ and $N$ the number of iterations. Second, we propose a dynamic optimal policy for the choice of the learning rate $(\gamma_k){k\geq 0}$ used in stochastic approximation (SA). We decompose our policy into two interacting levels: the inner and the outer level. In the inner level, we present the \nameref{Alg:v_4_s} algorithm (for "PAst Sign Search") which, based on a predefined sequence $(\gamma^o_k){k\geq 0}$, constructs a new sequence $(\gamma^i_k)_{k\geq 0}$ whose error decreases faster. In the outer level, we propose an optimal methodology for the selection of the predefined sequence $(\gamma^o_k)_{k\geq 0}$. Third, we show empirically that our selection methodology of the learning rate outperforms significantly standard algorithms used in reinforcement learning (RL) in the three following applications: the estimation of a drift, the optimal placement of limit orders and the optimal execution of large number of shares.