This work studies discrete-time discounted Markov decision processes with continuous state and action spaces and addresses the inverse problem of inferring a cost function from observed optimal behavior. We first consider the case in which we have access to the entire expert policy and characterize the set of solutions to the inverse problem by using occupation measures, linear duality, and complementary slackness conditions. To avoid trivial solutions and ill-posedness, we introduce a natural linear normalization constraint. This results in an infinite-dimensional linear feasibility problem, prompting a thorough analysis of its properties. Next, we use linear function approximators and adopt a randomized approach, namely the scenario approach and related probabilistic feasibility guarantees, to derive epsilon-optimal solutions for the inverse problem. We further discuss the sample complexity for a desired approximation accuracy. Finally, we deal with the more realistic case where we only have access to a finite set of expert demonstrations and a generative model and provide bounds on the error made when working with samples.

Naively trained Deep Reinforcement Learning agents may fail to satisfy vital safety constraints. To avoid costly retraining, we may desire to repair a previously trained reinforcement learning agent to obviate unsafe behaviour. We devise a counterexample-guided repair algorithm for repairing reinforcement learning systems leveraging safety critics. The algorithm jointly repairs a reinforcement learning agent and a safety critic using gradient-based constrained optimisation.

Linear Temporal Logic (LTL) offers a precise means for constraining the behavior of reinforcement learning agents. However, in many tasks, LTL is insufficient for task specification; LTL-constrained policy optimization, where the goal is to optimize a scalar reward under LTL constraints, is needed. Prior methods for this constrained problem are restricted to finite state spaces. In this work, we present Cycle Experience Replay (CyclER), a reward-shaping approach to this problem that allows continuous state and action spaces and the use of function approximations. CyclER guides a policy towards satisfaction by encouraging partial behaviors compliant with the LTL constraint, using the structure of the constraint. In doing so, it addresses the optimization challenges stemming from the sparse nature of LTL satisfaction. We evaluate CyclER in three continuous control domains. On these tasks, CyclER outperforms existing reward-shaping methods at finding performant and LTL-satisfying policies.

Q-learning excels in learning from feedback within sequential decision-making tasks but often requires extensive sampling to achieve significant improvements. While reward shaping can enhance learning efficiency, non-potential-based methods introduce biases that affect performance, and potential-based reward shaping, though unbiased, lacks the ability to provide heuristics for state-action pairs, limiting its effectiveness in complex environments. Large language models (LLMs) can achieve zero-shot learning for simpler tasks, but they suffer from low inference speeds and occasional hallucinations. To address these challenges, we propose \textbf{LLM-guided Q-learning}, a framework that leverages LLMs as heuristics to aid in learning the Q-function for reinforcement learning. Our theoretical analysis demonstrates that this approach adapts to hallucinations, improves sample efficiency, and avoids biasing final performance. Experimental results show that our algorithm is general, robust, and capable of preventing ineffective exploration.

This paper studies the transfer reinforcement learning (RL) problem where multiple RL problems have different reward functions but share the same underlying transition dynamics. In this setting, the Q-function of each RL problem (task) can be decomposed into a successor feature (SF) and a reward mapping: the former characterizes the transition dynamics, and the latter characterizes the task-specific reward function. This Q-function decomposition, coupled with a policy improvement operator known as generalized policy improvement (GPI), reduces the sample complexity of finding the optimal Q-function, and thus the SF \& GPI framework exhibits promising empirical performance compared to traditional RL methods like Q-learning. However, its theoretical foundations remain largely unestablished, especially when learning the successor features using deep neural networks (SF-DQN). This paper studies the provable knowledge transfer using SFs-DQN in transfer RL problems. We establish the first convergence analysis with provable generalization guarantees for SF-DQN with GPI. The theory reveals that SF-DQN with GPI outperforms conventional RL approaches, such as deep Q-network, in terms of both faster convergence rate and better generalization. Numerical experiments on real and synthetic RL tasks support the superior performance of SF-DQN \& GPI, aligning with our theoretical findings.

Reinforcement Learning (RL) suffers from sample inefficiency in sparse reward domains, and the problem is pronounced if there are stochastic transitions. To improve the sample efficiency, reward shaping is a well-studied approach to introduce intrinsic rewards that can help the RL agent converge to an optimal policy faster. However, designing a useful reward shaping function specific to each problem is challenging, even for domain experts. They would either have to rely on task-specific domain knowledge or provide an expert demonstration independently for each task. Given, that Large Language Models (LLMs) have rapidly gained prominence across a magnitude of natural language tasks, we aim to answer the following question: Can we leverage LLMs to construct a reward shaping function that can boost the sample efficiency of an RL agent? In this work, we aim to leverage off-the-shelf LLMs to generate a guide policy by solving a simpler deterministic abstraction of the original problem that can then be used to construct the reward shaping function for the downstream RL agent. Given the ineffectiveness of directly prompting LLMs, we propose MEDIC: a framework that augments LLMs with a Model-based feEDback critIC, which verifies LLM-generated outputs, to generate a possibly sub-optimal but valid plan for the abstract problem. Our experiments across domains from the BabyAI environment suite show 1) the effectiveness of augmenting LLMs with MEDIC, 2) a significant improvement in the sample complexity of PPO and A2C-based RL agents when guided by our LLM-generated plan, and finally, 3) pave the direction for further explorations of how these models can be used to augment existing RL pipelines.

Reinforcement Learning from Human Feedback (RLHF) has become the standard approach for aligning Large Language Models (LLMs) with human preferences, allowing LLMs to demonstrate remarkable abilities in various tasks. Existing methods work by emulating the preferences at the single decision (turn) level, limiting their capabilities in settings that require planning or multi-turn interactions to achieve a long-term goal. In this paper, we address this issue by developing novel methods for Reinforcement Learning (RL) from preference feedback between two full multi-turn conversations. In the tabular setting, we present a novel mirror-descent-based policy optimization algorithm for the general multi-turn preference-based RL problem, and prove its convergence to Nash equilibrium. To evaluate performance, we create a new environment, Education Dialogue, where a teacher agent guides a student in learning a random topic, and show that a deep RL variant of our algorithm outperforms RLHF baselines. Finally, we show that in an environment with explicit rewards, our algorithm recovers the same performance as a reward-based RL baseline, despite relying solely on a weaker preference signal.

Reinforcement Learning (RL)-Based Recommender Systems (RSs) have gained rising attention for their potential to enhance long-term user engagement. However, research in this field faces challenges, including the lack of user-friendly frameworks, inconsistent evaluation metrics, and difficulties in reproducing existing studies. To tackle these issues, we introduce EasyRL4Rec, an easy-to-use code library designed specifically for RL-based RSs. This library provides lightweight and diverse RL environments based on five public datasets and includes core modules with rich options, simplifying model development. It provides unified evaluation standards focusing on long-term outcomes and offers tailored designs for state modeling and action representation for recommendation scenarios. Furthermore, we share our findings from insightful experiments with current methods. EasyRL4Rec seeks to facilitate the model development and experimental process in the domain of RL-based RSs. The library is available for public use.

In this work, we consider data association problems involving multi-object tracking (MOT). In particular, we address the challenges arising from object occlusions. We propose a framework called approximate dynamic programming track (ADPTrack), which applies dynamic programming principles to improve an existing method called the base heuristic. Given a set of tracks and the next target frame, the base heuristic extends the tracks by matching them to the objects of this target frame directly. In contrast, ADPTrack first processes a few subsequent frames and applies the base heuristic starting from the next target frame to obtain tentative tracks. It then leverages the tentative tracks to match the objects of the target frame. This tends to reduce the occlusion-based errors and leads to an improvement over the base heuristic. When tested on the MOT17 video dataset, the proposed method demonstrates a 0.7% improvement in the association accuracy (IDF1 metric) over a state-of-the-art method that is used as the base heuristic. It also obtains improvements with respect to all the other standard metrics. Empirically, we found that the improvements are particularly pronounced in scenarios where the video data is obtained by fixed-position cameras.

We need to look at our shoelaces as we first learn to tie them but having mastered this skill, we can do it from touch alone. We call this phenomenon "sensory scaffolding": observation streams that are not needed by a master might yet aid a novice learner. We consider such sensory scaffolding setups for training artificial agents. For example, a robot arm may need to be deployed with just a low-cost, robust, general-purpose camera; yet its performance may improve by having privileged training-time-only access to informative albeit expensive and unwieldy motion capture rigs or fragile tactile sensors. For these settings, we propose Scaffolder, a reinforcement learning approach which effectively exploits privileged sensing in critics, world models, reward estimators, and other such auxiliary components that are only used at training time, to improve the target policy. For evaluating sensory scaffolding agents, we design a new "S3" suite of ten diverse simulated robotic tasks that explore a wide range of practical sensor setups. Agents must use privileged camera sensing to train blind hurdlers, privileged active visual perception to help robot arms overcome visual occlusions, privileged touch sensors to train robot hands, and more. Scaffolder easily outperforms relevant prior baselines and frequently performs comparably even to policies that have test-time access to the privileged sensors. It is well-known that Beethoven composed symphonies long after he had fully lost his hearing. Such feats are commonly held to be evidence of mastery: for example, novice typists need to look at the keyboard to locate keys but with practice, can graduate to typing without looking. Thus, sensing requirements may be different during learning versus after learning. We refer to this as "sensory scaffolding", drawing inspiration from the concept of scaffolding teaching mechanisms in psychology that provide temporary support for a student (Wood et al., 1976; Vygotsky et al., 2011), like training wheels when learning to ride a bicycle. For artificial learning agents such as robots, sensory scaffolding permits decoupling the observation streams required at test time from those that are used to train the agent. The sensors available in a deployed robot are often decided by practical considerations such as cost, robustness, size, compute requirements, and ease of instrumentation, e.g., autonomous cars with only cheap and robust RGB camera sensors. However, those considerations might carry less weight at training time, so a robot learning practitioner may choose to scaffold policy learning with privileged information (Vapnik & Vashist, 2009) from extra sensors available only at training. In the case of the cars above, the manufacturer might equip a small fleet of training cars with expensive privileged sensors like lidar to improve RGB-only driving policies for customers to install in their cars.