General ML · reinforcement learning

Inspired by the diversity and depth of XLand and the simplicity and minimalism of MiniGrid, we present XLand-MiniGrid, a suite of tools and grid-world environments for meta-reinforcement learning research. Written in JAX, XLand-MiniGrid is designed to be highly scalable and can potentially run on GPU or TPU accelerators, democratizing large-scale experimentation with limited resources. Along with the environments, XLand-MiniGrid provides pre-sampled benchmarks with millions of unique tasks of varying difficulty and easy-to-use baselines that allow users to quickly start training adaptive agents. In addition, we have conducted a preliminary analysis of scaling and generalization, showing that our baselines are capable of reaching millions of steps per second during training and validating that the proposed benchmarks are challenging. XLand-MiniGrid is open-source and available at \url{https://github.com/corl-team/xland-minigrid}.

We study a new class of MDPs that employs multinomial logit (MNL) function approximation to ensure valid probability distributions over the state space. Despite its significant benefits, incorporating the non-linear function raises substantial challenges in both *statistical* and *computational* efficiency. The best-known result of Hwang and Oh [2023] has achieved an $\widetilde{\mathcal{O}}(\kappa^{-1}dH^2\sqrt{K})$ regret upper bound, where $\kappa$ is a problem-dependent quantity, $d$ is the feature dimension, $H$ is the episode length, and $K$ is the number of episodes. However, we observe that $\kappa^{-1}$ exhibits polynomial dependence on the number of reachable states, which can be as large as the state space size in the worst case and thus undermines the motivation for function approximation. Additionally, their method requires storing all historical data and the time complexity scales linearly with the episode count, which is computationally expensive. In this work, we propose a statistically efficient algorithm that achieves a regret of $\widetilde{\mathcal{O}}(dH^2\sqrt{K} + \kappa^{-1}d^2H^2)$, eliminating the dependence on $\kappa^{-1}$ in the dominant term for the first time. We then address the computational challenges by introducing an enhanced algorithm that achieves the same regret guarantee but with only constant cost. Finally, we establish the first lower bound for this problem, justifying the optimality of our results in $d$ and $K$.

While originally developed for continuous control problems, Proximal Policy Optimization (PPO) has emerged as the work-horse of a variety of reinforcement learning (RL) applications, including the fine-tuning of generative models. Unfortunately, PPO requires multiple heuristics to enable stable convergence (e.g. value networks, clipping), and is notorious for its sensitivity to the precise implementation of these components. In response, we take a step back and ask what a *minimalist* RL algorithm for the era of generative models would look like. We propose REBEL, an algorithm that cleanly reduces the problem of policy optimization to regressing the *relative reward* between two completions to a prompt in terms of the policy, enabling strikingly lightweight implementation. In theory, we prove that fundamental RL algorithms like Natural Policy Gradient can be seen as variants of REBEL, which allows us to match the strongest known theoretical guarantees in terms of convergence and sample complexity in the RL literature. REBEL can also cleanly incorporate offline data and be extended to handle the intransitive preferences we frequently see in practice. Empirically, we find that REBEL provides a unified approach to language modeling and image generation with stronger or similar performance as PPO and DPO, all while being simpler to implement and more computationally efficient than PPO. When fine-tuning Llama-3-8B-Instruct, REBEL achieves strong performance in AlpacaEval 2.0, MT-Bench, and Open LLM Leaderboard. Implementation of REBEL can be found at <https://github.com/ZhaolinGao/REBEL>, and models trained by REBEL can be found at <https://huggingface.co/Cornell-AGI>.

We consider the problem of federated offline reinforcement learning (RL), a scenario under which distributed learning agents must collaboratively learn a high-quality control policy only using small pre-collected datasets generated according to different unknown behavior policies. Na\"{i}vely combining a standard offline RL approach with a standard federated learning approach to solve this problem can lead to poorly performing policies. In response, we develop the Federated Ensemble-Directed Offline Reinforcement Learning Algorithm (FEDORA), which distills the collective wisdom of the clients using an ensemble learning approach. We develop the FEDORA codebase to utilize distributed compute resources on a federated learning platform. We show that FEDORA significantly outperforms other approaches, including offline RL over the combined data pool, in various complex continuous control environments and real-world datasets. Finally, we demonstrate the performance of FEDORA in the real-world on a mobile robot. We provide our code and a video of our experiments at \url{https://github.com/DesikRengarajan/FEDORA}.

Curriculum Reinforcement Learning (CRL) is an approach to facilitate the learning process of agents by structuring tasks in a sequence of increasing complexity. Despite its potential, many existing CRL methods struggle to efficiently guide agents toward desired outcomes, particularly in the absence of domain knowledge. This paper introduces DiCuRL (Diffusion Curriculum Reinforcement Learning), a novel method that leverages conditional diffusion models to generate curriculum goals. To estimate how close an agent is to achieving its goal, our method uniquely incorporates a $Q$-function and a trainable reward function based on Adversarial Intrinsic Motivation within the diffusion model. Furthermore, it promotes exploration through the inherent noising and denoising mechanism present in the diffusion models and is environment-agnostic. This combination allows for the generation of challenging yet achievable goals, enabling agents to learn effectively without relying on domain knowledge. We demonstrate the effectiveness of DiCuRL in three different maze environments and two robotic manipulation tasks simulated in MuJoCo, where it outperforms or matches nine state-of-the-art CRL algorithms from the literature.

Among approaches for provably safe reinforcement learning, Model Predictive Shielding (MPS) has proven effective at complex tasks in continuous, high-dimensional state spaces, by leveraging a *backup policy* to ensure safety when the learned policy attempts to take risky actions. However, while MPS can ensure safety both during and after training, it often hinders task progress due to the conservative and task-oblivious nature of backup policies. This paper introduces *Dynamic Model Predictive Shielding* (DMPS), which optimizes reinforcement learning objectives while maintaining provable safety. DMPS employs a local planner to dynamically select safe recovery actions that maximize both short-term progress as well as long-term rewards. Crucially, the planner and the neural policy play a synergistic role in DMPS. When planning recovery actions for ensuring safety, the planner utilizes the neural policy to estimate long-term rewards, allowing it to *observe* beyond its short-term planning horizon. Conversely, the neural policy under training learns from the recovery plans proposed by the planner, converging to policies that are both *high-performing* and *safe* in practice. This approach guarantees safety during and after training, with bounded recovery regret that decreases exponentially with planning horizon depth. Experimental results demonstrate that DMPS converges to policies that rarely require shield interventions after training and achieve higher rewards compared to several state-of-the-art baselines.

We study reinforcement learning (RL) problems in which agents observe the reward or transition realizations at their current state _before deciding which action to take_. Such observations are available in many applications, including transactions, navigation and more. When the environment is known, previous work shows that this lookahead information can drastically increase the collected reward. However, outside of specific applications, existing approaches for interacting with unknown environments are not well-adapted to these observations. In this work, we close this gap and design provably-efficient learning algorithms able to incorporate lookahead information. To achieve this, we perform planning using the empirical distribution of the reward and transition observations, in contrast to vanilla approaches that only rely on estimated expectations. We prove that our algorithms achieve tight regret versus a baseline that also has access to lookahead information -- linearly increasing the amount of collected reward compared to agents that cannot handle lookahead information.

We introduce the Continuous Arcade Learning Environment (CALE), an extension of the well-known Arcade Learning Environment (ALE) [Bellemare et al., 2013]. The CALE uses the same underlying emulator of the Atari 2600 gaming system (Stella), but adds support for continuous actions. This enables the benchmarking and evaluation of continuous-control agents (such as PPO [Schulman et al., 2017] and SAC [Haarnoja et al., 2018]) and value-based agents (such as DQN [Mnih et al., 2015] and Rainbow [Hessel et al., 2018]) on the same environment suite. We provide a series of open questions and research directions that CALE enables, as well as initial baseline results using Soft Actor-Critic. CALE is available as part of the ALE athttps://github.com/Farama-Foundation/Arcade-Learning-Environment.

We focus on offline imitation learning (IL), which aims to mimic an expert's behavior using demonstrations without any interaction with the environment. One of the main challenges in offline IL is the limited support of expert demonstrations, which typically cover only a small fraction of the state-action space. While it may not be feasible to obtain numerous expert demonstrations, it is often possible to gather a larger set of sub-optimal demonstrations. For example, in treatment optimization problems, there are varying levels of doctor treatments available for different chronic conditions. These range from treatment specialists and experienced general practitioners to less experienced general practitioners. Similarly, when robots are trained to imitate humans in routine tasks, they might learn from individuals with different levels of expertise and efficiency. In this paper, we propose an offline IL approach that leverages the larger set of sub-optimal demonstrations while effectively mimicking expert trajectories. Existing offline IL methods based on behavior cloning or distribution matching often face issues such as overfitting to the limited set of expert demonstrations or inadvertently imitating sub-optimal trajectories from the larger dataset. Our approach, which is based on inverse soft-Q learning, learns from both expert and sub-optimal demonstrations. It assigns higher importance (through learned weights) to aligning with expert demonstrations and lower importance to aligning with sub-optimal ones. A key contribution of our approach, called SPRINQL, is transforming the offline IL problem into a convex optimization over the space of Q functions. Through comprehensive experimental evaluations, we demonstrate that the SPRINQL algorithm achieves state-of-the-art (SOTA) performance on offline IL benchmarks. Code is available at https://github.com/hmhuy0/SPRINQL .

Reinforcement learning (RL) traditionally trains agents using interaction data, which limits their capabilities to the scope of the training data. To create more knowledgeable agents, leveraging knowledge from large language models (LLMs) has shown a promising way. Despite various attempts to combine LLMs with RL, there is commonly a semantic gap between action signals and LLM tokens, which hinders their integration. This paper introduces a novel approach, KALM (Knowledgeable Agents from Language Model Rollouts), to learn knowledgeable agents by bridging this gap. KALM extracts knowledge from LLMs in the form of imaginary rollouts, which agents can learn through offline RL. To overcome the limitation that LLMs are inherently text-based and may be incompatible with numerical environmental data, KALM fine-tunes the LLM to perform bidirectional translation between textual goals and rollouts. This process enables the LLM to understand the environment better, facilitating the generation of meaningful rollouts. Experiments on robotic manipulation tasks demonstrate that KALM allows agents to rephrase complex goals and tackle novel tasks requiring new optimal behaviors. KALM achieves a 46% success rate in completing 1400 various novel goals, significantly outperforming the 26% success rate of baseline methods. Project homepage: https://kalmneurips2024.github.io.

The exploration-exploitation dilemma has been a central challenge in reinforcement learning (RL) with complex model classes. In this paper, we propose a new algorithm, Monotonic Q-Learning with Upper Confidence Bound (MQL-UCB) for RL with general function approximation. Our key algorithmic design includes (1) a general deterministic policy-switching strategy that achieves low switching cost, (2) a monotonic value function structure with carefully controlled function class complexity, and (3) a variance-weighted regression scheme that exploits historical trajectories with high data efficiency. MQL-UCB achieves minimax optimal regret of $\tilde{O}(d\sqrt{HK})$ when $K$ is sufficiently large and near-optimal policy switching cost of $\tilde{O}(dH)$, with $d$ being the eluder dimension of the function class, $H$ being the planning horizon, and $K$ being the number of episodes. Our work sheds light on designing provably sample-efficient and deployment-efficient Q-learning with nonlinear function approximation.

We study the sample complexity of learning an $\varepsilon$-optimal policy in an average-reward Markov decision process (MDP) under a generative model. For weakly communicating MDPs, we establish the complexity bound $\widetilde{O}\left(SA\frac{\mathsf{H}}{\varepsilon^2} \right)$, where $\mathsf{H}$ is the span of the bias function of the optimal policy and $SA$ is the cardinality of the state-action space. Our result is the first that is minimax optimal (up to log factors) in all parameters $S,A,\mathsf{H}$, and $\varepsilon$, improving on existing work that either assumes uniformly bounded mixing times for all policies or has suboptimal dependence on the parameters. We also initiate the study of sample complexity in general (multichain) average-reward MDPs. We argue a new transient time parameter $\mathsf{B}$ is necessary, establish an $\widetilde{O}\left(SA\frac{\mathsf{B} + \mathsf{H}}{\varepsilon^2} \right)$ complexity bound, and prove a matching (up to log factors) minimax lower bound. Both results are based on reducing the average-reward MDP to a discounted MDP, which requires new ideas in the general setting. To optimally analyze this reduction, we develop improved bounds for $\gamma$-discounted MDPs, showing that $\widetilde{O}\left(SA\frac{\mathsf{H}}{(1-\gamma)^2\varepsilon^2} \right)$ and $\widetilde{O}\left(SA\frac{\mathsf{B} + \mathsf{H}}{(1-\gamma)^2\varepsilon^2} \right)$ samples suffice to learn $\varepsilon$-optimal policies in weakly communicating and in general MDPs, respectively. Both these results circumvent the well-known minimax lower bound of $\widetilde{\Omega}\left(SA\frac{1}{(1-\gamma)^3\varepsilon^2} \right)$ for $\gamma$-discounted MDPs, and establish a quadratic rather than cubic horizon dependence for a fixed MDP instance.

Partial observability of the underlying states generally presents significant challenges for reinforcement learning (RL). In practice, certain *privileged information* , e.g., the access to states from simulators, has been exploited in training and achieved prominent empirical successes. To better understand the benefits of privileged information, we revisit and examine several simple and practically used paradigms in this setting, with both computation and sample efficiency analyses. Specifically, we first formalize the empirical paradigm of *expert distillation* (also known as *teacher-student* learning), demonstrating its pitfall in finding near-optimal policies. We then identify a condition of the partially observable environment, the deterministic filter condition, under which expert distillation achieves sample and computational complexities that are *both* polynomial. Furthermore, we investigate another successful empirical paradigm of *asymmetric actor-critic*, and focus on the more challenging setting of observable partially observable Markov decision processes. We develop a belief-weighted optimistic asymmetric actor-critic algorithm with polynomial sample and quasi-polynomial computational complexities, where one key component is a new provable oracle for learning belief states that preserve *filter stability* under a misspecified model, which may be of independent interest. Finally, we also investigate the provable efficiency of partially observable multi-agent RL (MARL) with privileged information. We develop algorithms with the feature of centralized-training-with-decentralized-execution, a popular framework in empirical MARL, with polynomial sample and (quasi-)polynomial computational complexity in both paradigms above. Compared with a few recent related theoretical studies, our focus is on understanding practically inspired algorithmic paradigms, without computationally intractable oracles.

While imitation learning requires access to high-quality data, offline reinforcement learning (RL) should, in principle, perform similarly or better with substantially lower data quality by using a value function. However, current results indicate that offline RL often performs worse than imitation learning, and it is often unclear what holds back the performance of offline RL. Motivated by this observation, we aim to understand the bottlenecks in current offline RL algorithms. While poor performance of offline RL is typically attributed to an imperfect value function, we ask: *is the main bottleneck of offline RL indeed in learning the value function, or something else?* To answer this question, we perform a systematic empirical study of (1) value learning, (2) policy extraction, and (3) policy generalization in offline RL problems, analyzing how these components affect performance. We make two surprising observations. First, we find that the choice of a policy extraction algorithm significantly affects the performance and scalability of offline RL, often more so than the value learning objective. For instance, we show that common value-weighted behavioral cloning objectives (e.g., AWR) do not fully leverage the learned value function, and switching to behavior-constrained policy gradient objectives (e.g., DDPG+BC) often leads to substantial improvements in performance and scalability. Second, we find that a big barrier to improving offline RL performance is often imperfect policy generalization on test-time states out of the support of the training data, rather than policy learning on in-distribution states. We then show that the use of suboptimal but high-coverage data or test-time policy training techniques can address this generalization issue in practice. Specifically, we propose two simple test-time policy improvement methods and show that these methods lead to better performance.

Procedurally generated environments such as Procgen Benchmark provide a testbed for evaluating the agent's ability to robustly learn a relevant skill, by situating the agent in ever-changing levels. The diverse levels associated with varying contexts are naturally connected to curriculum learning. Existing works mainly focus on arranging the levels to explicitly form a curriculum. In this work, we take a close look at the learning process itself under the multi-level training in Procgen. Interestingly, the learning process exhibits a gradual shift from easy contexts to hard contexts, suggesting an implicit curriculum in multi-level training. Our analysis is made possible through C-Procgen, a benchmark we build upon Procgen that enables explicit control of the contexts. We believe our findings will foster a deeper understanding of learning in diverse contexts, and our benchmark will benefit future research in curriculum reinforcement learning.

Memory models such as Recurrent Neural Networks (RNNs) and Transformers address Partially Observable Markov Decision Processes (POMDPs) by mapping trajectories to latent Markov states. Neither model scales particularly well to long sequences, especially compared to an emerging class of memory models called Linear Recurrent Models. We discover that the recurrent update of these models resembles a monoid, leading us to reformulate existing models using a novel monoid-based framework that we call memoroids. We revisit the traditional approach to batching in recurrent reinforcement learning, highlighting theoretical and empirical deficiencies. We leverage memoroids to propose a batching method that improves sample efficiency, increases the return, and simplifies the implementation of recurrent loss functions in reinforcement learning.

Inverse reinforcement learning (IRL) aims to learn a reward function and a corresponding policy that best fit the demonstrated trajectories of an expert. However, current IRL works cannot learn incrementally from an ongoing trajectory because they have to wait to collect at least one complete trajectory to learn. To bridge the gap, this paper considers the problem of learning a reward function and a corresponding policy while observing the initial state-action pair of an ongoing trajectory and keeping updating the learned reward and policy when new state-action pairs of the ongoing trajectory are observed. We formulate this problem as an online bi-level optimization problem where the upper level dynamically adjusts the learned reward according to the newly observed state-action pairs with the help of a meta-regularization term, and the lower level learns the corresponding policy. We propose a novel algorithm to solve this problem and guarantee that the algorithm achieves sub-linear local regret $O(\sqrt{T}+\log T+\sqrt{T}\log T)$. If the reward function is linear, we prove that the proposed algorithm achieves sub-linear regret $O(\log T)$. Experiments are used to validate the proposed algorithm.

Reinforcement learning (RL) algorithms are typically based on optimizing a Markov Decision Process (MDP) using the optimal Bellman equation. Recent studies have revealed that focusing the optimization of Bellman equations solely on in-sample actions tends to result in more stable optimization, especially in the presence of function approximation. Upon on these findings, in this paper, we propose an Empirical MDP Iteration (EMIT) framework. EMIT constructs a sequence of empirical MDPs using data from the growing replay memory. For each of these empirical MDPs, it learns an estimated Q-function denoted as $\widehat{Q}$. The key strength is that by restricting the Bellman update to in-sample bootstrapping, each empirical MDP converges to a unique optimal $\widehat{Q}$ function. Furthermore, gradually expanding from the empirical MDPs to the original MDP induces a monotonic policy improvement. Instead of creating entirely new algorithms, we demonstrate that EMIT can be seamlessly integrated with existing online RL algorithms, effectively acting as a regularizer for contemporary Q-learning methods. We show this by implementing EMIT for two representative RL algorithms, DQN and TD3. Experimental results on Atari and MuJoCo benchmarks show that EMIT significantly reduces estimation errors and substantially improves the performance of both algorithms.

Existing Maximum-Entropy (MaxEnt) Reinforcement Learning (RL) methods for continuous action spaces are typically formulated based on actor-critic frameworks and optimized through alternating steps of policy evaluation and policy improvement. In the policy evaluation steps, the critic is updated to capture the soft Q-function. In the policy improvement steps, the actor is adjusted in accordance with the updated soft Q-function. In this paper, we introduce a new MaxEnt RL framework modeled using Energy-Based Normalizing Flows (EBFlow). This framework integrates the policy evaluation steps and the policy improvement steps, resulting in a single objective training process. Our method enables the calculation of the soft value function used in the policy evaluation target without Monte Carlo approximation. Moreover, this design supports the modeling of multi-modal action distributions while facilitating efficient action sampling. To evaluate the performance of our method, we conducted experiments on the MuJoCo benchmark suite and a number of high-dimensional robotic tasks simulated by Omniverse Isaac Gym. The evaluation results demonstrate that our method achieves superior performance compared to widely-adopted representative baselines.

We consider offline imitation learning (IL), which aims to train an agent to imitate from the dataset of expert demonstrations without online interaction with the environment. Behavioral Cloning (BC) has been a simple yet effective approach to offline IL, but it is also well-known to be vulnerable to the covariate shift resulting from the mismatch between the state distributions induced by the learned policy and the expert policy. Moreover, as often occurs in practice, when expert datasets are collected from an arbitrary state distribution instead of a stationary one, these shifts become more pronounced, potentially leading to substantial failures in existing IL methods. Specifically, we focus on covariate shift resulting from arbitrary state data distributions, such as biased data collection or incomplete trajectories, rather than shifts induced by changes in dynamics or noisy expert actions. In this paper, to mitigate the effect of the covariate shifts in BC, we propose DrilDICE, which utilizes a distributionally robust BC objective by employing a stationary distribution correction ratio estimation (DICE) to derive a feasible solution. We evaluate the effectiveness of our method through an extensive set of experiments covering diverse covariate shift scenarios. The results demonstrate the efficacy of the proposed approach in improving the robustness against the shifts, outperforming existing offline IL methods in such scenarios.