We derive a new analysis of Follow The Regularized Leader (FTRL) for online learning with delayed bandit feedback. By separating the cost of delayed feedback from that of bandit feedback, our analysis allows us to obtain new results in three important settings. On the one hand, we derive the first optimal (up to logarithmic factors) regret bounds for combinatorial semi-bandits with delay and adversarial Markov decision processes with delay (and known transition functions). On the other hand, we use our analysis to derive an efficient algorithm for linear bandits with delay achieving near-optimal regret bounds. Our novel regret decomposition shows that FTRL remains stable across multiple rounds under mild assumptions on the Hessian of the regularizer.
Policy Optimization (PO) is one of the most popular methods in Reinforcement Learning (RL). Thus, theoretical guarantees for PO algorithms have become especially important to the RL community. In this paper, we study PO in adversarial MDPs with a challenge that arises in almost every real-world application -- \textit{delayed bandit feedback}. We give the first near-optimal regret bounds for PO in tabular MDPs, and may even surpass state-of-the-art (which uses less efficient methods). Our novel Delay-Adapted PO (DAPO) is easy to implement and to generalize, allowing us to extend our algorithm to: (i) infinite state space under the assumption of linear $Q$-function, proving the first regret bounds for delayed feedback with function approximation. (ii) deep RL, demonstrating its effectiveness in experiments on MuJoCo domains.
Policy optimization is among the most popular and successful reinforcement learning algorithms, and there is increasing interest in understanding its theoretical guarantees. In this work, we initiate the study of policy optimization for the stochastic shortest path (SSP) problem, a goal-oriented reinforcement learning model that strictly generalizes the finite-horizon model and better captures many applications. We consider a wide range of settings, including stochastic and adversarial environments under full information or bandit feedback, and propose a policy optimization algorithm for each setting that makes use of novel correction terms and/or variants of dilated bonuses (Luo et al., 2021). For most settings, our algorithm is shown to achieve a near-optimal regret bound. One key technical contribution of this work is a new approximation scheme to tackle SSP problems that we call \textit{stacked discounted approximation} and use in all our proposed algorithms. Unlike the finite-horizon approximation that is heavily used in recent SSP algorithms, our new approximation enables us to learn a near-stationary policy with only logarithmic changes during an episode and could lead to an exponential improvement in space complexity.
The standard assumption in reinforcement learning (RL) is that agents observe feedback for their actions immediately. However, in practice feedback is often observed in delay. This paper studies online learning in episodic Markov decision process (MDP) with unknown transitions, adversarially changing costs, and unrestricted delayed bandit feedback. More precisely, the feedback for the agent in episode $k$ is revealed only in the end of episode $k + d^k$, where the delay $d^k$ can be changing over episodes and chosen by an oblivious adversary. We present the first algorithms that achieve near-optimal $\sqrt{K + D}$ regret, where $K$ is the number of episodes and $D = \sum_{k=1}^K d^k$ is the total delay, significantly improving upon the best known regret bound of $(K + D)^{2/3}$.
We study cooperative online learning in stochastic and adversarial Markov decision process (MDP). That is, in each episode, $m$ agents interact with an MDP simultaneously and share information in order to minimize their individual regret. We consider environments with two types of randomness: \emph{fresh} -- where each agent's trajectory is sampled i.i.d, and \emph{non-fresh} -- where the realization is shared by all agents (but each agent's trajectory is also affected by its own actions). More precisely, with non-fresh randomness the realization of every cost and transition is fixed at the start of each episode, and agents that take the same action in the same state at the same time observe the same cost and next state. We thoroughly analyze all relevant settings, highlight the challenges and differences between the models, and prove nearly-matching regret lower and upper bounds. To our knowledge, we are the first to consider cooperative reinforcement learning (RL) with either non-fresh randomness or in adversarial MDPs.
The classical Policy Iteration (PI) algorithm alternates between greedy one-step policy improvement and policy evaluation. Recent literature shows that multi-step lookahead policy improvement leads to a better convergence rate at the expense of increased complexity per iteration. However, prior to running the algorithm, one cannot tell what is the best fixed lookahead horizon. Moreover, per a given run, using a lookahead of horizon larger than one is often wasteful. In this work, we propose for the first time to dynamically adapt the multi-step lookahead horizon as a function of the state and of the value estimate. We devise two PI variants and analyze the trade-off between iteration count and computational complexity per iteration. The first variant takes the desired contraction factor as the objective and minimizes the per-iteration complexity. The second variant takes as input the computational complexity per iteration and minimizes the overall contraction factor. We then devise a corresponding DQN-based algorithm with an adaptive tree search horizon. We also include a novel enhancement for on-policy learning: per-depth value function estimator. Lastly, we demonstrate the efficacy of our adaptive lookahead method in a maze environment and in Atari.
We study the Stochastic Shortest Path (SSP) problem in which an agent has to reach a goal state in minimum total expected cost. In the learning formulation of the problem, the agent has no prior knowledge about the costs and dynamics of the model. She repeatedly interacts with the model for $K$ episodes, and has to learn to approximate the optimal policy as closely as possible. In this work we show that the minimax regret for this setting is $\widetilde O(B_\star \sqrt{|S| |A| K})$ where $B_\star$ is a bound on the expected cost of the optimal policy from any state, $S$ is the state space, and $A$ is the action space. This matches the lower bound of Rosenberg et al. (2020) up to logarithmic factors, and improves their regret bound by a factor of $\sqrt{|S|}$. Our algorithm runs in polynomial-time per episode, and is based on a novel reduction to reinforcement learning in finite-horizon MDPs. To that end, we provide an algorithm for the finite-horizon setting whose leading term in the regret depends only logarithmically on the horizon, yielding the same regret guarantees for SSP.
Reinforcement learning typically assumes that the agent observes feedback from the environment immediately, but in many real-world applications (like recommendation systems) the feedback is observed in delay. Thus, we consider online learning in episodic Markov decision processes (MDPs) with unknown transitions, adversarially changing costs and unrestricted delayed feedback. That is, the costs and trajectory of episode $k$ are only available at the end of episode $k + d^k$, where the delays $d^k$ are neither identical nor bounded, and are chosen by an adversary. We present novel algorithms based on policy optimization that achieve near-optimal high-probability regret of $\widetilde O ( \sqrt{K} + \sqrt{D} )$ under full-information feedback, where $K$ is the number of episodes and $D = \sum_{k} d^k$ is the total delay. Under bandit feedback, we prove similar $\widetilde O ( \sqrt{K} + \sqrt{D} )$ regret assuming that the costs are stochastic, and $\widetilde O ( K^{2/3} + D^{2/3} )$ regret in the general case. To our knowledge, we are the first to consider the important setting of delayed feedback in adversarial MDPs.
We consider provably-efficient reinforcement learning (RL) in non-episodic factored Markov decision processes (FMDPs). All previous algorithms for regret minimization in this setting made the strong assumption that the factored structure of the FMDP is known to the learner in advance. In this paper, we provide the first provably-efficient algorithm that has to learn the structure of the FMDP while minimizing its regret. Our algorithm is based on the optimism in face of uncertainty principle, combined with a simple statistical method for structure learning, and can be implemented efficiently given oracle-access to an FMDP planner. It maintains its computational efficiency even though the number of possible structures is exponential.