We prove Wasserstein inverse reinforcement learning enables the learner's reward values to imitate the expert's reward values in a finite iteration for multi-objective optimizations. Moreover, we prove Wasserstein inverse reinforcement learning enables the learner's optimal solutions to imitate the expert's optimal solutions for multi-objective optimizations with lexicographic order.

Solving job shop scheduling problems (JSSPs) with a fixed strategy, such as a priority dispatching rule, may yield satisfactory results for several problem instances but, nevertheless, insufficient results for others. From this single-strategy perspective finding a near optimal solution to a specific JSSP varies in difficulty even if the machine setup remains the same. A recent intensively researched and promising method to deal with difficulty variability is Deep Reinforcement Learning (DRL), which dynamically adjusts an agent's planning strategy in response to difficult instances not only during training, but also when applied to new situations. In this paper, we further improve DLR as an underlying method by actively incorporating the variability of difficulty within the same problem size into the design of the learning process. We base our approach on a state-of-the-art methodology that solves JSSP by means of DRL and graph neural network embeddings. Our work supplements the training routine of the agent by a curriculum learning strategy that ranks the problem instances shown during training by a new metric of problem instance difficulty. Our results show that certain curricula lead to significantly better performances of the DRL solutions. Agents trained on these curricula beat the top performance of those trained on randomly distributed training data, reaching 3.2% shorter average makespans.

Reinforcement Learning (RL) with constraints is becoming an increasingly important problem for various applications. Often, the average criterion is more suitable than the discounted criterion. Yet, RL for average criterion-constrained MDPs remains a challenging problem. Algorithms designed for discounted constrained RL problems often do not perform well for the average CMDP setting. In this paper, we introduce a new policy optimization with function approximation algorithm for constrained MDPs with the average criterion. The Average-Constrained Policy Optimization (ACPO) algorithm is inspired by the famed PPO-type algorithms based on trust region methods. We develop basic sensitivity theory for average MDPs, and then use the corresponding bounds in the design of the algorithm. We provide theoretical guarantees on its performance, and through extensive experimental work in various challenging MuJoCo environments, show the superior performance of the algorithm when compared to other state-of-the-art algorithms adapted for the average CMDP setting.

Interest in reinforcement learning (RL) has recently surged due to the application of deep learning techniques, but these connectionist approaches are opaque compared with symbolic systems. Learning Classifier Systems (LCSs) are evolutionary machine learning systems that can be categorised as eXplainable AI (XAI) due to their rule-based nature. Michigan LCSs are commonly used in RL domains as the alternative Pittsburgh systems (e.g. SAMUEL) suffer from complex algorithmic design and high computational requirements; however they can produce more compact/interpretable solutions than Michigan systems. We aim to develop two novel Pittsburgh LCSs to address RL domains: PPL-DL and PPL-ST. The former acts as a "zeroth-level" system, and the latter revisits SAMUEL's core Monte Carlo learning mechanism for estimating rule strength. We compare our two Pittsburgh systems to the Michigan system XCS across deterministic and stochastic FrozenLake environments. Results show that PPL-ST performs on-par or better than PPL-DL and outperforms XCS in the presence of high levels of environmental uncertainty. Rulesets evolved by PPL-ST can achieve higher performance than those evolved by XCS, but in a more parsimonious and therefore more interpretable fashion, albeit with higher computational cost. This indicates that PPL-ST is an LCS well-suited to producing explainable policies in RL domains.

The discovery of individual objectives in collective behavior of complex dynamical systems such as fish schools and bacteria colonies is a long-standing challenge. Inverse reinforcement learning is a potent approach for addressing this challenge but its applicability to dynamical systems, involving continuous state-action spaces and multiple interacting agents, has been limited. In this study, we tackle this challenge by introducing an off-policy inverse multi-agent reinforcement learning algorithm (IMARL). Our approach combines the ReF-ER techniques with guided cost learning. By leveraging demonstrations, our algorithm automatically uncovers the reward function and learns an effective policy for the agents. Through extensive experimentation, we demonstrate that the proposed policy captures the behavior observed in the provided data, and achieves promising results across problem domains including single agent models in the OpenAI gym and multi-agent models of schooling behavior. The present study shows that the proposed IMARL algorithm is a significant step towards understanding collective dynamics from the perspective of its constituents, and showcases its value as a tool for studying complex physical systems exhibiting collective behaviour.

In the realm of urban transportation, metro systems serve as crucial and sustainable means of public transit. However, their substantial energy consumption poses a challenge to the goal of sustainability. Disturbances such as delays and passenger flow changes can further exacerbate this issue by negatively affecting energy efficiency in metro systems. To tackle this problem, we propose a policy-based reinforcement learning approach that reschedules the metro timetable and optimizes energy efficiency in metro systems under disturbances by adjusting the dwell time and cruise speed of trains. Our experiments conducted in a simulation environment demonstrate the superiority of our method over baseline methods, achieving a traction energy consumption reduction of up to 10.9% and an increase in regenerative braking energy utilization of up to 47.9%. This study provides an effective solution to the energy-saving problem of urban rail transit.

As a pivotal component to attaining generalizable solutions in human intelligence, reasoning provides great potential for reinforcement learning (RL) agents' generalization towards varied goals by summarizing part-to-whole arguments and discovering cause-and-effect relations. However, how to discover and represent causalities remains a huge gap that hinders the development of causal RL. In this paper, we augment Goal-Conditioned RL (GCRL) with Causal Graph (CG), a structure built upon the relation between objects and events. We novelly formulate the GCRL problem into variational likelihood maximization with CG as latent variables. To optimize the derived objective, we propose a framework with theoretical performance guarantees that alternates between two steps: using interventional data to estimate the posterior of CG; using CG to learn generalizable models and interpretable policies. Due to the lack of public benchmarks that verify generalization capability under reasoning, we design nine tasks and then empirically show the effectiveness of the proposed method against five baselines on these tasks. Further theoretical analysis shows that our performance improvement is attributed to the virtuous cycle of causal discovery, transition modeling, and policy training, which aligns with the experimental evidence in extensive ablation studies.

This paper studies tabular reinforcement learning (RL) in the hybrid setting, which assumes access to both an offline dataset and online interactions with the unknown environment. A central question boils down to how to efficiently utilize online data collection to strengthen and complement the offline dataset and enable effective policy fine-tuning. Leveraging recent advances in reward-agnostic exploration and model-based offline RL, we design a three-stage hybrid RL algorithm that beats the best of both worlds -- pure offline RL and pure online RL -- in terms of sample complexities. The proposed algorithm does not require any reward information during data collection. Our theory is developed based on a new notion called single-policy partial concentrability, which captures the trade-off between distribution mismatch and miscoverage and guides the interplay between offline and online data.

Despite remarkable successes, deep reinforcement learning algorithms remain sample inefficient: they require an enormous amount of trial and error to find good policies. Model-based algorithms promise sample efficiency by building an environment model that can be used for planning. Posterior Sampling for Reinforcement Learning is such a model-based algorithm that has attracted significant interest due to its performance in the tabular setting. This paper introduces Posterior Sampling for Deep Reinforcement Learning (PSDRL), the first truly scalable approximation of Posterior Sampling for Reinforcement Learning that retains its model-based essence. PSDRL combines efficient uncertainty quantification over latent state space models with a specially tailored continual planning algorithm based on value-function approximation. Extensive experiments on the Atari benchmark show that PSDRL significantly outperforms previous state-of-the-art attempts at scaling up posterior sampling while being competitive with a state-of-the-art (model-based) reinforcement learning method, both in sample efficiency and computational efficiency.

We show the convergence of Wasserstein inverse reinforcement learning for multi-objective optimizations with the projective subgradient method by formulating an inverse problem of the multi-objective optimization problem. In addition, we prove convergence of inverse reinforcement learning (maximum entropy inverse reinforcement learning, guided cost learning) with gradient descent and the projective subgradient method.