Inspired by the recent successes of Inverse Optimization (IO) across various application domains, we propose a novel offline Reinforcement Learning (ORL) algorithm for continuous state and action spaces, leveraging the convex loss function called ``sub-optimality loss" from the IO literature. To mitigate the distribution shift commonly observed in ORL problems, we further employ a robust and non-causal Model Predictive Control (MPC) expert steering a nominal model of the dynamics using in-hindsight information stemming from the model mismatch. Unlike the existing literature, our robust MPC expert enjoys an exact and tractable convex reformulation. In the second part of this study, we show that the IO hypothesis class, trained by the proposed convex loss function, enjoys ample expressiveness and achieves competitive performance comparing with the state-of-the-art (SOTA) methods in the low-data regime of the MuJoCo benchmark while utilizing three orders of magnitude fewer parameters, thereby requiring significantly fewer computational resources. To facilitate the reproducibility of our results, we provide an open-source package implementing the proposed algorithms and the experiments.

Reinforcement learning (RL) is a powerful approach for robot learning. However, model-free RL (MFRL) requires a large number of environment interactions to learn successful control policies. This is due to the noisy RL training updates and the complexity of robotic systems, which typically involve highly non-linear dynamics and noisy sensor signals. In contrast, model-based RL (MBRL) not only trains a policy but simultaneously learns a world model that captures the environment's dynamics and rewards. The world model can either be used for planning, for data collection, or to provide first-order policy gradients for training. Leveraging a world model significantly improves sample efficiency compared to model-free RL. However, training a world model alongside the policy increases the computational complexity, leading to longer training times that are often intractable for complex real-world scenarios. In this work, we propose a new method for accelerating model-based RL using state-space world models. Our approach leverages state-space models (SSMs) to parallelize the training of the dynamics model, which is typically the main computational bottleneck. Additionally, we propose an architecture that provides privileged information to the world model during training, which is particularly relevant for partially observable environments. We evaluate our method in several real-world agile quadrotor flight tasks, involving complex dynamics, for both fully and partially observable environments. We demonstrate a significant speedup, reducing the world model training time by up to 10 times, and the overall MBRL training time by up to 4 times. This benefit comes without compromising performance, as our method achieves similar sample efficiency and task rewards to state-of-the-art MBRL methods.

Reinforcement Learning (RL) agents have demonstrated their potential across various robotic tasks. However, they still heavily rely on human-engineered reward functions, requiring extensive trial-and-error and access to target behavior information, often unavailable in real-world settings. This paper introduces REDS: REward learning from Demonstration with Segmentations, a novel reward learning framework that leverages action-free videos with minimal supervision. Specifically, REDS employs video demonstrations segmented into subtasks from diverse sources and treats these segments as ground-truth rewards. We train a dense reward function conditioned on video segments and their corresponding subtasks to ensure alignment with ground-truth reward signals by minimizing the Equivalent-Policy Invariant Comparison distance. Additionally, we employ contrastive learning objectives to align video representations with subtasks, ensuring precise subtask inference during online interactions. Our experiments show that REDS significantly outperforms baseline methods on complex robotic manipulation tasks in Meta-World and more challenging real-world tasks, such as furniture assembly in FurnitureBench, with minimal human intervention. Moreover, REDS facilitates generalization to unseen tasks and robot embodiments, highlighting its potential for scalable deployment in diverse environments.

With the increasingly congested and contested space environment, safe and effective satellite operation has become increasingly challenging. As a result, there is growing interest in autonomous satellite capabilities, with common machine learning techniques gaining attention for their potential to address complex decision-making in the space domain. However, the "black-box" nature of many of these methods results in difficulty understanding the model's input/output relationship and more specifically its sensitivity to environmental disturbances, sensor noise, and control intervention. This paper explores the use of Deep Reinforcement Learning (DRL) for satellite control in multi-agent inspection tasks. The Local Intelligent Network of Collaborative Satellites (LINCS) Lab is used to test the performance of these control algorithms across different environments, from simulations to real-world quadrotor UAV hardware, with a particular focus on understanding their behavior and potential degradation in performance when deployed beyond the training environment.

Reinforcement learning (RL) post-training is crucial for LLM alignment and reasoning, but existing policy-based methods, such as PPO and DPO, can fall short of fixing shortcuts inherited from pre-training. In this work, we introduce $Q\sharp$, a value-based algorithm for KL-regularized RL that guides the reference policy using the optimal regularized $Q$ function. We propose to learn the optimal $Q$ function using distributional RL on an aggregated online dataset. Unlike prior value-based baselines that guide the model using unregularized $Q$-values, our method is theoretically principled and provably learns the optimal policy for the KL-regularized RL problem. Empirically, $Q\sharp$ outperforms prior baselines in math reasoning benchmarks while maintaining a smaller KL divergence to the reference policy. Theoretically, we establish a reduction from KL-regularized RL to no-regret online learning, providing the first bounds for deterministic MDPs under only realizability. Thanks to distributional RL, our bounds are also variance-dependent and converge faster when the reference policy has small variance. In sum, our results highlight $Q\sharp$ as an effective approach for post-training LLMs, offering both improved performance and theoretical guarantees. The code can be found at https://github.com/jinpz/q_sharp.

Improving sampling efficiency and generalization capability is critical for the successful data-driven control of quadrotor unmanned aerial vehicles (UAVs) that are inherently unstable. While various reinforcement learning (RL) approaches have been applied to autonomous quadrotor flight, they often require extensive training data, posing multiple challenges and safety risks in practice. To address these issues, we propose data-efficient, equivariant monolithic and modular RL frameworks for quadrotor low-level control. Specifically, by identifying the rotational and reflectional symmetries in quadrotor dynamics and encoding these symmetries into equivariant network models, we remove redundancies of learning in the state-action space. This approach enables the optimal control action learned in one configuration to automatically generalize into other configurations via symmetry, thereby enhancing data efficiency. Experimental results demonstrate that our equivariant approaches significantly outperform their non-equivariant counterparts in terms of learning efficiency and flight performance.

Reinforcement learning has delivered promising results in achieving human- or even superhuman-level capabilities across diverse problem domains, but success in dexterous robot manipulation remains limited. This work investigates the key challenges in applying reinforcement learning to solve a collection of contact-rich manipulation tasks on a humanoid embodiment. We introduce novel techniques to overcome the identified challenges with empirical validation. Our main contributions include an automated real-to-sim tuning module that brings the simulated environment closer to the real world, a generalized reward design scheme that simplifies reward engineering for long-horizon contact-rich manipulation tasks, a divide-and-conquer distillation process that improves the sample efficiency of hard-exploration problems while maintaining sim-to-real performance, and a mixture of sparse and dense object representations to bridge the sim-to-real perception gap. We show promising results on three humanoid dexterous manipulation tasks, with ablation studies on each technique. Our work presents a successful approach to learning humanoid dexterous manipulation using sim-to-real reinforcement learning, achieving robust generalization and high performance without the need for human demonstration.

Stable and robust robotic grasping is essential for current and future robot applications. In recent works, the use of large datasets and supervised learning has enhanced speed and precision in antipodal grasping. However, these methods struggle with perception and calibration errors due to large planning horizons. To obtain more robust and reactive grasping motions, leveraging reinforcement learning combined with tactile sensing is a promising direction. Yet, there is no systematic evaluation of how the complexity of force-based tactile sensing affects the learning behavior for grasping tasks. This paper compares various tactile and environmental setups using two model-free reinforcement learning approaches for antipodal grasping. Our findings suggest that under imperfect visual perception, various tactile features improve learning outcomes, while complex tactile inputs complicate training.

Artificial Intelligence increasingly pursues large, complex models that perform many tasks within increasingly realistic domains. How, if at all, should these developments in AI influence cognitive science? We argue that progress in AI offers timely opportunities for cognitive science to embrace experiments with increasingly naturalistic stimuli, tasks, and behaviors; and computational models that can accommodate these changes. We first review a growing body of research spanning neuroscience, cognitive science, and AI that suggests that incorporating a broader range of naturalistic experimental paradigms (and models that accommodate them) may be necessary to resolve some aspects of natural intelligence and ensure that our theories generalize. We then suggest that integrating recent progress in AI and cognitive science will enable us to engage with more naturalistic phenomena without giving up experimental control or the pursuit of theoretically grounded understanding. We offer practical guidance on how methodological practices can contribute to cumulative progress in naturalistic computational cognitive science, and illustrate a path towards building computational models that solve the real problems of natural cognition - together with a reductive understanding of the processes and principles by which they do so.

Reinforcement learning algorithms typically necessitate extensive exploration of the state space to find optimal policies. However, in safety-critical applications, the risks associated with such exploration can lead to catastrophic consequences. Existing safe exploration methods attempt to mitigate this by imposing constraints, which often result in overly conservative behaviours and inefficient learning. Heavy penalties for early constraint violations can trap agents in local optima, deterring exploration of risky yet high-reward regions of the state space. To address this, we introduce a method that explicitly learns state-conditioned safety representations. By augmenting the state features with these safety representations, our approach naturally encourages safer exploration without being excessively cautious, resulting in more efficient and safer policy learning in safety-critical scenarios. Empirical evaluations across diverse environments show that our method significantly improves task performance while reducing constraint violations during training, underscoring its effectiveness in balancing exploration with safety.