This paper investigates how to incorporate expert observations (without explicit information on expert actions) into a deep reinforcement learning setting to improve sample efficiency. First, we formulate an augmented policy loss combining a maximum entropy reinforcement learning objective with a behavioral cloning loss that leverages a forward dynamics model. Then, we propose an algorithm that automatically adjusts the weights of each component in the augmented loss function. Experiments on a variety of continuous control tasks demonstrate that the proposed algorithm outperforms various benchmarks by effectively utilizing available expert observations.

Underpinning the past decades of work on the design, initialization, and optimization of neural networks is a seemingly innocuous assumption: that the network is trained on a \textit{stationary} data distribution. In settings where this assumption is violated, e.g.\ deep reinforcement learning, learning algorithms become unstable and brittle with respect to hyperparameters and even random seeds. One factor driving this instability is the loss of plasticity, meaning that updating the network's predictions in response to new information becomes more difficult as training progresses. While many recent works provide analyses and partial solutions to this phenomenon, a fundamental question remains unanswered: to what extent do known mechanisms of plasticity loss overlap, and how can mitigation strategies be combined to best maintain the trainability of a network? This paper addresses these questions, showing that loss of plasticity can be decomposed into multiple independent mechanisms and that, while intervening on any single mechanism is insufficient to avoid the loss of plasticity in all cases, intervening on multiple mechanisms in conjunction results in highly robust learning algorithms. We show that a combination of layer normalization and weight decay is highly effective at maintaining plasticity in a variety of synthetic nonstationary learning tasks, and further demonstrate its effectiveness on naturally arising nonstationarities, including reinforcement learning in the Arcade Learning Environment.

Reinforcement learning (RL) [43] has emerged as a powerful framework for sequential decision-making in dynamic and uncertain environments. While traditional RL methods, predominantly focused on maximizing the expected return, have seen significant advancements through approaches such as Q-learning [37, 25] and policy gradients [28, 10], they often fall short in real-world scenarios demanding strict risk control, such as financial investment [9], medical treatment [16], and automous driving [11]. The significance of comprehending risk management in RL has led to the emergence of Risk-Sensitive RL (RSRL). Unlike risk-neutral RL, which primarily focuses on maximizing expected returns, RSRL seeks to optimize risk metrics, such as entropy risk measures (ERM) [17, 18] or conditional value-at-risk (CVaR) [46], of the possible cumulative reward which emphasizes its distributional characteristics. However, traditional RL framework based on Q-learning which typically considers the mean of reward-to-go and corresponding Bellman equation, cannot efficiently capture the characteristics of the cumulative reward's distribution. Therefore, there has been an upsurge of interest in Distributional RL (DisRL) due to its capacity to understand the intrinsic distributional attributes of cumulative rewards, which has already achieved significant empirical success in risk-sensitive tasks [8, 14, 30, 45, 34].

The integration of Unmanned Aerial Vehicles (UAVs) into Search and Rescue (SAR) missions presents a promising avenue for enhancing operational efficiency and effectiveness. However, the success of these missions is not solely dependent on the technical capabilities of the drones but also on their acceptance and interaction with humans on the ground. This paper explores the effect of human-centric factor in UAV trajectory planning for SAR missions. We introduce a novel approach based on the reinforcement learning augmented with Analytic Hierarchy Process and novel similarity-based experience replay to optimize UAV trajectories, balancing operational objectives with human comfort and safety considerations. Additionally, through a comprehensive survey, we investigate the impact of gender cues and anthropomorphism in UAV design on public acceptance and trust, revealing significant implications for drone interaction strategies in SAR. Our contributions include (1) a reinforcement learning framework for UAV trajectory planning that dynamically integrates multi-objective considerations, (2) an analysis of human perceptions towards gendered and anthropomorphized drones in SAR contexts, and (3) the application of similarity-based experience replay for enhanced learning efficiency in complex SAR scenarios. The findings offer valuable insights into designing UAV systems that are not only technically proficient but also aligned with human-centric values.

In order for autonomous mobile robots to navigate in human spaces, they must abide by our social norms. Reinforcement learning (RL) has emerged as an effective method to train sequential decision-making policies that are able to respect these norms. However, a large portion of existing work in the field conducts both RL training and testing in simplistic environments. This limits the generalization potential of these models to unseen environments, and the meaningfulness of their reported results. We propose a method to improve the generalization performance of RL social navigation methods using curriculum learning. By employing multiple environment types and by modeling pedestrians using multiple dynamics models, we are able to progressively diversify and escalate difficulty in training. Our results show that the use of curriculum learning in training can be used to achieve better generalization performance than previous training methods. We also show that results presented in many existing state-of-the-art RL social navigation works do not evaluate their methods outside of their training environments, and thus do not reflect their policies' failure to adequately generalize to out-of-distribution scenarios. In response, we validate our training approach on larger and more crowded testing environments than those used in training, allowing for more meaningful measurements of model performance.

This work pioneers regret analysis of risk-sensitive reinforcement learning in partially observable environments with hindsight observation, addressing a gap in theoretical exploration. We introduce a novel formulation that integrates hindsight observations into a Partially Observable Markov Decision Process (POMDP) framework, where the goal is to optimize accumulated reward under the entropic risk measure. We develop the first provably efficient RL algorithm tailored for this setting. We also prove by rigorous analysis that our algorithm achieves polynomial regret $\tilde{O}\left(\frac{e^{|{\gamma}|H}-1}{|{\gamma}|H}H^2\sqrt{KHS^2OA}\right)$, which outperforms or matches existing upper bounds when the model degenerates to risk-neutral or fully observable settings. We adopt the method of change-of-measure and develop a novel analytical tool of beta vectors to streamline mathematical derivations. These techniques are of particular interest to the theoretical study of reinforcement learning.

Abstract--This paper introduces a Multi-Agent Deep Reinforcement Learning (MA-DRL) approach for routing in Low Earth Orbit Satellite Constellations (LSatCs). Each satellite is an independent decision-making agent with a partial knowledge of the environment, and supported by feedback received from the nearby agents. Building on our previous work that introduced a Q-routing solution, the contribution of this paper is to extend it to a deep learning framework able to quickly adapt to the network and traffic changes, and based on two phases: (1) An offline exploration learning phase that relies on a global Deep Neural Network (DNN) to learn the optimal paths at each possible position and congestion level; (2) An online exploitation phase with local, on-board, pre-trained DNNs. Results show that MA-DRL efficiently learns optimal routes offline that are then loaded for an efficient distributed routing online. Low Earth Orbit (LEO) Satellite Constellations (LSatCs) are one of the pillars of 6G ubiquitous and global connectivity, enhancing cellular coverage, supporting a global backbone, and enabling advanced applications [1].

Offline safe reinforcement learning (RL) aims to train a constraint satisfaction policy from a fixed dataset. Current state-of-the-art approaches are based on supervised learning with a conditioned policy. However, these approaches fall short in real-world applications that involve complex tasks with rich temporal and logical structures. In this paper, we propose temporal logic Specification-conditioned Decision Transformer (SDT), a novel framework that harnesses the expressive power of signal temporal logic (STL) to specify complex temporal rules that an agent should follow and the sequential modeling capability of Decision Transformer (DT). Empirical evaluations on the DSRL benchmarks demonstrate the better capacity of SDT in learning safe and high-reward policies compared with existing approaches. In addition, SDT shows good alignment with respect to different desired degrees of satisfaction of the STL specification that it is conditioned on.

Energy Internet (EI) is emerging as new share economy platform for flexible local energy supplies in smart cities. Empowered by the Internet-of-Things (IoT) and Artificial Intelligence (AI), EI aims to unlock peer-to-peer energy trading and sharing among prosumers, who can adeptly switch roles between providers and consumers in localized energy markets with rooftop photovoltaic panels, vehicle-to-everything technologies, packetized energy management, etc. The integration of prosumers in EI, however, will encounter many challenges in modelling, analyzing, and designing an efficient, economic, and social-optimal platform for energy sharing, calling for advanced AI/IoT-based solutions to resource optimization, information exchange, and interaction protocols in the context of the share economy. In this study, we aim to introduce a recently emerged paradigm, Machina Economicus, to investigate the economic rationality in modelling, analysis, and optimization of AI/IoT-based EI prosumer behaviors. The new paradigm, built upon the theory of machine learning and mechanism design, will offer new angles to investigate the selfishness of AI through a game-theoretic perspective, revealing potential competition and collaborations resulting from the self-adaptive learning and decision-making capacity. This study will focus on how the introduction of AI will reshape prosumer behaviors on the EI, and how this paradigm will reveal new research questions and directions when AI meets the share economy. With an extensive case analysis in the literature, we will also shed light on potential solutions for advancements of AI in future smart cities.

Network systems form the foundation of modern society, playing a critical role in various applications. However, these systems are at significant risk of being adversely affected by unforeseen circumstances, such as disasters. Considering this, there is a pressing need for research to enhance the robustness of network systems. Recently, in reinforcement learning, the relationship between acquiring robustness and regularizing entropy has been identified. Additionally, imitation learning is used within this framework to reflect experts' behavior. However, there are no comprehensive studies on the use of a similar imitation framework for optimal transport on networks. Therefore, in this study, imitation-regularized optimal transport (I-OT) on networks was investigated. It encodes prior knowledge on the network by imitating a given prior distribution. The I-OT solution demonstrated robustness in terms of the cost defined on the network. Moreover, we applied the I-OT to a logistics planning problem using real data. We also examined the imitation and apriori risk information scenarios to demonstrate the usefulness and implications of the proposed method.