Intelligent energy management strategies, such as Vehicle-to-Grid (V2G) and Grid-to-Vehicle (G2V) emerge as a potential solution to the Electric Vehicles' (EVs) integration into the energy grid. These strategies promise enhanced grid resilience and economic benefits for both vehicle owners and grid operators. Despite the announced prospective, the adoption of these strategies is still hindered by an array of operational problems. Key among these is the lack of a simulation platform that allows to validate and refine V2G and G2V strategies. Including the development, training, and testing in the context of Energy Communities (ECs) incorporating multiple flexible energy assets. Addressing this gap, first we introduce the EVLearn, a simulation module for researching in both V2G and G2V energy management strategies, that models EVs, their charging infrastructure and associated energy flexibility dynamics; second, this paper integrates EVLearn with the existing CityLearn framework, providing V2G and G2V simulation capabilities into the study of broader energy management strategies. Results validated EVLearn and its integration into CityLearn, where the impact of these strategies is highlighted through a comparative simulation scenario.

Abstract-- In this study, a chair-type asymmetric tripedal low-rigidity robot was designed based on the three-legged chair character in the movie "Suzume" and its gait was generated. Its body structure consists of three legs that are asymmetric to the body, so it cannot be easily balanced. In addition, the actuator is a servo motor that can only feed-forward rotational angle commands and the sensor can only sense the robot's posture quaternion. In such an asymmetric and imperfect body structure, we analyzed how gait is generated in walking and stand-up motions by generating gaits with two different methods: a method using linear completion to connect the postures necessary for the gait discovered through trial and error using the actual robot, and a method using the gait generated by reinforcement learning in the simulator and reflecting it to the actual robot. Both methods were able to generate gait that realized walking and stand-up motions, and interesting gait patterns were observed, which differed depending on the method, and were confirmed on the actual robot.

Background: Clinical diagnosis is typically reached by following a series of steps recommended by guidelines authored by colleges of experts. Accordingly, guidelines play a crucial role in rationalizing clinical decisions but suffer from limitations as they are built to cover the majority of the population and fail at covering patients with uncommon conditions. Moreover, their updates are long and expensive, making them unsuitable for emerging diseases and practices. Methods: Inspired by guidelines, we formulate the task of diagnosis as a sequential decision-making problem and study the use of Deep Reinforcement Learning (DRL) algorithms to learn the optimal sequence of actions to perform in order to obtain a correct diagnosis from Electronic Health Records (EHRs). We apply DRL on synthetic, but realistic EHRs and develop two clinical use cases: Anemia diagnosis, where the decision pathways follow the schema of a decision tree; and Systemic Lupus Erythematosus (SLE) diagnosis, which follows a weighted criteria score. We particularly evaluate the robustness of our approaches to noisy and missing data since these frequently occur in EHRs. Results: In both use cases, and in the presence of imperfect data, our best DRL algorithms exhibit competitive performance when compared to the traditional classifiers, with the added advantage that they enable the progressive generation of a pathway to the suggested diagnosis which can both guide and explain the decision-making process. Conclusion: DRL offers the opportunity to learn personalized decision pathways to diagnosis. We illustrate with our two use cases their advantages: they generate step-by-step pathways that are self-explanatory; and their correctness is competitive when compared to state-of-the-art approaches.

Autonomous navigation capabilities play a critical role in service robots operating in environments where human interactions are pivotal, due to the dynamic and unpredictable nature of these environments. However, the variability in human behavior presents a substantial challenge for robots in predicting and anticipating movements, particularly in crowded scenarios. To address this issue, a memory-enabled deep reinforcement learning framework is proposed for autonomous robot navigation in diverse pedestrian scenarios. The proposed framework leverages long-term memory to retain essential information about the surroundings and model sequential dependencies effectively. The importance of human-robot interactions is also encoded to assign higher attention to these interactions. A global planning mechanism is incorporated into the memory-enabled architecture. Additionally, a multi-term reward system is designed to prioritize and encourage long-sighted robot behaviors by incorporating dynamic warning zones. Simultaneously, it promotes smooth trajectories and minimizes the time taken to reach the robot's desired goal. Extensive simulation experiments show that the suggested approach outperforms representative state-of-the-art methods, showcasing its ability to a navigation efficiency and safety in real-world scenarios.

Markov decision processes (MDPs) are a fundamental model for decision making under uncertainty. They exhibit non-deterministic choice as well as probabilistic uncertainty. Traditionally, verification algorithms assume exact knowledge of the probabilities that govern the behaviour of an MDP. As this assumption is often unrealistic in practice, statistical model checking (SMC) was developed in the past two decades. It allows to analyse MDPs with unknown transition probabilities and provide probably approximately correct (PAC) guarantees on the result. Model-based SMC algorithms sample the MDP and build a model of it by estimating all transition probabilities, essentially for every transition answering the question: ``What are the odds?'' However, so far the statistical methods employed by the state of the art SMC algorithms are quite naive. Our contribution are several fundamental improvements to those methods: On the one hand, we survey statistics literature for better concentration inequalities; on the other hand, we propose specialised approaches that exploit our knowledge of the MDP. Our improvements are generally applicable to many kinds of problem statements because they are largely independent of the setting. Moreover, our experimental evaluation shows that they lead to significant gains, reducing the number of samples that the SMC algorithm has to collect by up to two orders of magnitude.

Multi-task reinforcement learning (MTRL) demonstrate potential for enhancing the generalization of a robot, enabling it to perform multiple tasks concurrently. However, the performance of MTRL may still be susceptible to conflicts between tasks and negative interference. To facilitate efficient MTRL, we propose Task-Specific Action Correction (TSAC), a general and complementary approach designed for simultaneous learning of multiple tasks. TSAC decomposes policy learning into two separate policies: a shared policy (SP) and an action correction policy (ACP). To alleviate conflicts resulting from excessive focus on specific tasks' details in SP, ACP incorporates goal-oriented sparse rewards, enabling an agent to adopt a long-term perspective and achieve generalization across tasks. Additional rewards transform the original problem into a multi-objective MTRL problem. Furthermore, to convert the multi-objective MTRL into a single-objective formulation, TSAC assigns a virtual expected budget to the sparse rewards and employs Lagrangian method to transform a constrained single-objective optimization into an unconstrained one. Experimental evaluations conducted on Meta-World's MT10 and MT50 benchmarks demonstrate that TSAC outperforms existing state-of-the-art methods, achieving significant improvements in both sample efficiency and effective action execution.

This paper analyzes multi-step TD-learning algorithms within the `deadly triad' scenario, characterized by linear function approximation, off-policy learning, and bootstrapping. In particular, we prove that n-step TD-learning algorithms converge to a solution as the sampling horizon n increases sufficiently. The paper is divided into two parts. In the first part, we comprehensively examine the fundamental properties of their model-based deterministic counterparts, including projected value iteration, gradient descent algorithms, and the control theoretic approach, which can be viewed as prototype deterministic algorithms whose analysis plays a pivotal role in understanding and developing their model-free reinforcement learning counterparts. In particular, we prove that these algorithms converge to meaningful solutions when n is sufficiently large. Based on these findings, two n-step TD-learning algorithms are proposed and analyzed, which can be seen as the model-free reinforcement learning counterparts of the gradient and control theoretic algorithms.

Understanding the transition events between metastable states in complex systems is an important subject in the fields of computational physics, chemistry and biology. The transition pathway plays an important role in characterizing the mechanism underlying the transition, for example, in the study of conformational changes of bio-molecules. In fact, computing the transition pathway is a challenging task for complex and high-dimensional systems. In this work, we formulate the path-finding task as a cost minimization problem over a particular path space. The cost function is adapted from the Freidlin-Wentzell action functional so that it is able to deal with rough potential landscapes. The path-finding problem is then solved using a actor-critic method based on the deep deterministic policy gradient algorithm (DDPG). The method incorporates the potential force of the system in the policy for generating episodes and combines physical properties of the system with the learning process for molecular systems. The exploitation and exploration nature of reinforcement learning enables the method to efficiently sample the transition events and compute the globally optimal transition pathway. We illustrate the effectiveness of the proposed method using three benchmark systems including an extended Mueller system and the Lennard-Jones system of seven particles.

Large Vision-Language Models (LVLMs) have demonstrated proficiency in tackling a variety of visual-language tasks. However, current LVLMs suffer from misalignment between text and image modalities which causes three kinds of hallucination problems, i.e., object existence, object attribute, and object relationship. To tackle this issue, existing methods mainly utilize Reinforcement Learning (RL) to align modalities in LVLMs. However, they still suffer from three main limitations: (1) General feedback can not indicate the hallucination type contained in the response; (2) Sparse rewards only give the sequence-level reward for the whole response; and (3)Annotation cost is time-consuming and labor-intensive. To handle these limitations, we propose an innovative method to align modalities in LVLMs through Fine-Grained Artificial Intelligence Feedback (FGAIF), which mainly consists of three steps: AI-based Feedback Collection, Fine-grained Reward Model Training, and Reinforcement Learning with Fine-grained Reward. Specifically, We first utilize AI tools to predict the types of hallucination for each segment in the response and obtain a collection of fine-grained feedback. Then, based on the collected reward data, three specialized reward models are trained to produce dense rewards. Finally, a novel fine-grained feedback module is integrated into the Proximal Policy Optimization (PPO) algorithm. Extensive experiments are conducted on hallucination and general benchmarks, demonstrating the superior performance of our proposed method. Notably, compared with previous models trained with the RL-based aligning method, our proposed method is effective even with fewer parameters.

Surrogate rewards for linear temporal logic (LTL) objectives are commonly utilized in planning problems for LTL objectives. In a widely-adopted surrogate reward approach, two discount factors are used to ensure that the expected return approximates the satisfaction probability of the LTL objective. The expected return then can be estimated by methods using the Bellman updates such as reinforcement learning. However, the uniqueness of the solution to the Bellman equation with two discount factors has not been explicitly discussed. We demonstrate with an example that when one of the discount factors is set to one, as allowed in many previous works, the Bellman equation may have multiple solutions, leading to inaccurate evaluation of the expected return. We then propose a condition for the Bellman equation to have the expected return as the unique solution, requiring the solutions for states inside a rejecting bottom strongly connected component (BSCC) to be 0. We prove this condition is sufficient by showing that the solutions for the states with discounting can be separated from those for the states without discounting under this condition.