Abstract--Cyber-Physical Systems (CPS) often leverage Reinforcement Learning (RL) techniques to adapt dynamically to changing environments and optimize performance. However, it is challenging to construct safety cases for RL components. We therefore propose the SAFE-RL (Safety and Accountability Framework for Evaluating Reinforcement Learning) for supporting the development, validation, and safe deployment of RLbased CPS. We adopt a design science approach to construct the framework and demonstrate its use in three RL applications in small Uncrewed Aerial systems (sUAS). Reinforcement Learning (RL) has become a key enabler of self-adaptive behaviors in Cyber-Physical Systems (CPS). Each ASK node adapting to dynamic and uncertain environments, RL enhances shows an icon for its related trustworthiness dimensions.

In the context of global urbanization and motorization, traffic congestion has become a significant issue, severely affecting the quality of life, environment, and economy. This paper puts forward a single-agent reinforcement learning (RL)-based regional traffic signal control (TSC) model. Different from multi - agent systems, this model can coordinate traffic signals across a large area, with the goals of alleviating regional traffic congestion and minimizing the total travel time. The TSC environment is precisely defined through specific state space, action space, and reward functions. The state space consists of the current congestion state, which is represented by the queue lengths of each link, and the current signal phase scheme of intersections. The action space is designed to select an intersection first and then adjust its phase split. Two reward functions are meticulously crafted. One focuses on alleviating congestion and the other aims to minimize the total travel time while considering the congestion level. The experiments are carried out with the SUMO traffic simulation software. The performance of the TSC model is evaluated by comparing it with a base case where no signal-timing adjustments are made. The results show that the model can effectively control congestion. For example, the queuing length is significantly reduced in the scenarios tested. Moreover, when the reward is set to both alleviate congestion and minimize the total travel time, the average travel time is remarkably decreased, which indicates that the model can effectively improve traffic conditions. This research provides a new approach for large-scale regional traffic signal control and offers valuable insights for future urban traffic management.

This work presents the MarineGym, a high-performance reinforcement learning (RL) platform specifically designed for underwater robotics. It aims to address the limitations of existing underwater simulation environments in terms of RL compatibility, training efficiency, and standardized benchmarking. MarineGym integrates a proposed GPU-accelerated hydrodynamic plugin based on Isaac Sim, achieving a rollout speed of 250,000 frames per second on a single NVIDIA RTX 3060 GPU. It also provides five models of unmanned underwater vehicles (UUVs), multiple propulsion systems, and a set of predefined tasks covering core underwater control challenges. Additionally, the DR toolkit allows flexible adjustments of simulation and task parameters during training to improve Sim2Real transfer. Further benchmark experiments demonstrate that MarineGym improves training efficiency over existing platforms and supports robust policy adaptation under various perturbations. We expect this platform could drive further advancements in RL research for underwater robotics. For more details about MarineGym and its applications, please visit our project page: https://marine-gym.com/.

We present a hierarchical neuro-symbolic control framework that couples classical symbolic planning with transformer-based policies to address complex, long-horizon decision-making tasks. At the high level, a symbolic planner constructs an interpretable sequence of operators based on logical propositions, ensuring systematic adherence to global constraints and goals. At the low level, each symbolic operator is translated into a sub-goal token that conditions a decision transformer to generate a fine-grained sequence of actions in uncertain, high-dimensional environments. We provide theoretical analysis showing how approximation errors from both the symbolic planner and the neural execution layer accumulate. Empirical evaluations in grid-worlds with multiple keys, locked doors, and item-collection tasks show that our hierarchical approach outperforms purely end-to-end neural approach in success rates and policy efficiency.

Multi-task reinforcement learning (MTRL) aims to endow a single agent with the ability to perform well on multiple tasks. Recent works have focused on developing novel sophisticated architectures to improve performance, often resulting in larger models; it is unclear, however, whether the performance gains are a consequence of the architecture design itself or the extra parameters. We argue that gains are mostly due to scale by demonstrating that naïvely scaling up a simple MTRL baseline to match parameter counts outperforms the more sophisticated architectures, and these gains benefit most from scaling the critic over the actor. Additionally, we explore the training stability advantages that come with task diversity, demonstrating that increasing the number of tasks can help mitigate plasticity loss. Our findings suggest that MTRL's simultaneous training across multiple tasks provides a natural framework for beneficial parameter scaling in reinforcement learning, challenging the need for complex architectural innovations.

Reinforcement learning (RL)-based fine-tuning has emerged as a powerful approach for aligning diffusion models with black-box objectives. Proximal policy optimization (PPO) is the most popular choice of method for policy optimization. While effective in terms of performance, PPO is highly sensitive to hyper-parameters and involves substantial computational overhead. REINFORCE, on the other hand, mitigates some computational complexities such as high memory overhead and sensitive hyper-parameter tuning, but has suboptimal performance due to high-variance and sample inefficiency. While the variance of the REINFORCE can be reduced by sampling multiple actions per input prompt and using a baseline correction term, it still suffers from sample inefficiency. To address these challenges, we systematically analyze the efficiency-effectiveness trade-off between REINFORCE and PPO, and propose leave-one-out PPO (LOOP), a novel RL for diffusion fine-tuning method. LOOP combines variance reduction techniques from REINFORCE, such as sampling multiple actions per input prompt and a baseline correction term, with the robustness and sample efficiency of PPO via clipping and importance sampling. Our results demonstrate that LOOP effectively improves diffusion models on various black-box objectives, and achieves a better balance between computational efficiency and performance.

Real-world sequential decision-making tasks often require balancing trade-offs between multiple conflicting objectives, making Multi-Objective Reinforcement Learning (MORL) an increasingly prominent field of research. Despite recent advances, existing MORL literature has narrowly focused on performance within static environments, neglecting the importance of generalizing across diverse settings. Conversely, existing research on generalization in RL has always assumed scalar rewards, overlooking the inherent multi-objectivity of real-world problems. Generalization in the multi-objective context is fundamentally more challenging, as it requires learning a Pareto set of policies addressing varying preferences across multiple objectives. In this paper, we formalize the concept of generalization in MORL and how it can be evaluated. We then contribute a novel benchmark featuring diverse multi-objective domains with parameterized environment configurations to facilitate future studies in this area. Our baseline evaluations of state-of-the-art MORL algorithms on this benchmark reveals limited generalization capabilities, suggesting significant room for improvement. Our empirical findings also expose limitations in the expressivity of scalar rewards, emphasizing the need for multi-objective specifications to achieve effective generalization. We further analyzed the algorithmic complexities within current MORL approaches that could impede the transfer in performance from the single- to multiple-environment settings. This work fills a critical gap and lays the groundwork for future research that brings together two key areas in reinforcement learning: solving multi-objective decision-making problems and generalizing across diverse environments. We make our code available at https://github.com/JaydenTeoh/MORL-Generalization.

Practical control systems pose significant challenges in identifying optimal control policies due to uncertainties in the system model and external disturbances. While $H_\infty$ control techniques are commonly used to design robust controllers that mitigate the effects of disturbances, these methods often require complex and computationally intensive calculations. To address this issue, this paper proposes a reinforcement learning algorithm called Robust Deterministic Policy Gradient (RDPG), which formulates the $H_\infty$ control problem as a two-player zero-sum dynamic game. In this formulation, one player (the user) aims to minimize the cost, while the other player (the adversary) seeks to maximize it. We then employ deterministic policy gradient (DPG) and its deep reinforcement learning counterpart to train a robust control policy with effective disturbance attenuation. In particular, for practical implementation, we introduce an algorithm called robust deep deterministic policy gradient (RDDPG), which employs a deep neural network architecture and integrates techniques from the twin-delayed deep deterministic policy gradient (TD3) to enhance stability and learning efficiency. To evaluate the proposed algorithm, we implement it on an unmanned aerial vehicle (UAV) tasked with following a predefined path in a disturbance-prone environment. The experimental results demonstrate that the proposed method outperforms other control approaches in terms of robustness against disturbances, enabling precise real-time tracking of moving targets even under severe disturbance conditions.

Abstract--Mobile Edge Computing (MEC) offers low-latency and high-bandwidth support for Internet-of-Vehicles (IoV) applications. However, due to high vehicle mobility and finite communication coverage of base stations, it is hard to maintain uninterrupted and high-quality services without proper service migration among MEC servers. Existing solutions commonly rely on prior knowledge and rarely consider efficient resource allocation during the service migration process, making it hard to reach optimal performance in dynamic IoV environments. To address these important challenges, we propose SR-CL, a novel mobility-aware seamless Service migration and Resource allocation framework via Convex-optimization-enabled deep reinforcement Learning in multi-edge IoV systems. First, we decouple the Mixed Integer Nonlinear Programming (MINLP) problem of service migration and resource allocation into two sub-problems. Next, we design a new actor-critic-based asynchronous-update deep reinforcement learning method to handle service migration, where the delayed-update actor makes migration decisions and the one-step-update critic evaluates the decisions to guide the policy update. Notably, we theoretically derive the optimal resource allocation with convex optimization for each MEC server, thereby further improving system performance. Using the real-world datasets of vehicle trajectories and testbed, extensive experiments are conducted to verify the effectiveness of the proposed SR-CL. Compared to benchmark methods, the SR-CL achieves superior convergence and delay performance under various scenarios. However, the real-time demands of IoV applications pose When vehicles offload tasks, MEC servers create dedicated significant challenges for onboard processors with limited service instances via virtualization techniques for the vehicles computational capabilities [2]. Although Cloud Computing and allocate proper resources to them [7].

Single-player contextual bandits are a well-studied problem in reinforcement learning that has seen applications in various fields such as advertising, healthcare, and finance. In light of the recent work on \emph{information asymmetric} bandits \cite{chang2022online, chang2023online}, we propose a novel multiplayer information asymmetric contextual bandit framework where there are multiple players each with their own set of actions. At every round, they observe the same context vectors and simultaneously take an action from their own set of actions, giving rise to a joint action. However, upon taking this action the players are subjected to information asymmetry in (1) actions and/or (2) rewards. We designed an algorithm \texttt{LinUCB} by modifying the classical single-player algorithm \texttt{LinUCB} in \cite{chu2011contextual} to achieve the optimal regret $O(\sqrt{T})$ when only one kind of asymmetry is present. We then propose a novel algorithm \texttt{ETC} that is built on explore-then-commit principles to achieve the same optimal regret when both types of asymmetry are present.