Long-horizon tasks in robotic manipulation present significant challenges in reinforcement learning (RL) due to the difficulty of designing dense reward functions and effectively exploring the expansive state-action space. However, despite a lack of dense rewards, these tasks often have a multi-stage structure, which can be leveraged to decompose the overall objective into manageable subgoals. In this work, we propose DEMO3, a framework that exploits this structure for efficient learning from visual inputs. Specifically, our approach incorporates multi-stage dense reward learning, a bi-phasic training scheme, and world model learning into a carefully designed demonstration-augmented RL framework that strongly mitigates the challenge of exploration in long-horizon tasks. Our evaluations demonstrate that our method improves data-efficiency by an average of 40% and by 70% on particularly difficult tasks compared to state-of-the-art approaches. We validate this across 16 sparse-reward tasks spanning four domains, including challenging humanoid visual control tasks using as few as five demonstrations.

--Hybrid light fidelity (LiFi) and wireless fidelity (WiFi) indoor networks has been envisioned as a promising technology to alleviate radio frequency spectrum crunch to accommodate the ever-increasing data rate demand in indoor scenarios. The hybrid LiFi/WiFi indoor networks can leverage the advantages of fast data transmission from LiFi and wider coverage of WiFi, thus complementing well with each other and further improving the network performance compared with the standalone networks. However, to leverage the co-existence, several challenges should be addressed, including but not limited to user association, mobility support, and efficient resource allocation. Therefore, the objective of the paper is to design a new user-access point association algorithm to maximize the sum throughput of the hybrid networks. We first mathematically formulate the sum data rate maximization problem by determining the AP selection for each user in indoor networks with consideration of user mobility and practical capacity limitations, which is a nonconvex binary integer programming problem. T o solve this problem, we then propose a sequential-proximal policy optimization (S-PPO) based deep reinforcement learning method. Extensive simulations are conducted to evaluate the proposed method by comparing it with exhaustive search (ES), signal strength strategy (SSS), and trust region policy optimization (TRPO) methods. Comprehensive simulation results demonstrate that our solution algorithm can outperform SSS by about 32.25% of the sum throughput and 19.09% of the fairness on average, and outperform TRPO by about 10.34% and 10.23%, respectively. Over the past few years, the usage of the internet has been continuously increasing. According to the latest data, people spend an average of 6 hours and 58 minutes daily on screens connected to the internet [1]. Moreover, an increasing number of applications require high-speed support, such as video calls, VR gaming, streaming media, and so on. However, we are facing a global digital divide, i.e., internet speeds in urban areas are often much faster than in rural areas, due to the generally less developed internet infrastructure in rural locations. Visible light communication (VLC), where light-emitting diodes (LEDs) can be used to transmit data by optical spectrum, has been envisioned as a promising solution for last-mile access because of its high bandwidth, enhanced security, electromagnetic interference-free nature, and easy integration with existing infrastructure [2]-[7].

Reinforcement learning (RL) offers powerful techniques for solving complex sequential decision-making tasks from experience. In this paper, we demonstrate how RL can be applied to adversarial machine learning (AML) to develop a new class of attacks that learn to generate adversarial examples: inputs designed to fool machine learning models. Unlike traditional AML methods that craft adversarial examples independently, our RL-based approach retains and exploits past attack experience to improve future attacks. We formulate adversarial example generation as a Markov Decision Process and evaluate RL's ability to (a) learn effective and efficient attack strategies and (b) compete with state-of-the-art AML. On CIFAR-10, our agent increases the success rate of adversarial examples by 19.4% and decreases the median number of victim model queries per adversarial example by 53.2% from the start to the end of training. In a head-to-head comparison with a state-of-the-art image attack, SquareAttack, our approach enables an adversary to generate adversarial examples with 13.1% more success after 5000 episodes of training. From a security perspective, this work demonstrates a powerful new attack vector that uses RL to attack ML models efficiently and at scale.

Reinforcement learning in psychology (as opposed to machine learning) refers to a family of mathematical models of how animals and humans learn. It has its origins with Thorndike's Law of Effect: behavior with positive outcomes is reinforced and likely to be repeated (learned). Reinforcement learning is part of a larger family of stochastic learning models where behavior is probabilistic (Bush and Mosteller 1951, 1953, 1955). The key ideas are that the STATE OF LEARNING of a SUBJECT (person or animal) is represented by a vector in a STATE SPACE. The subject's behavior (or RESPONSE) given a STIMULUS is not deterministic, but depends on probabilities determined by the state of learning. The OUTCOME(or PAYOFF) changes the state of learning. In reinforcement learning, the relative size of the payoff determines how strongly (if at all) the behavior is reinforced.

This work examines the question: How can a robot explore efficiently? We conduct a human user study to gain insights into effective exploration strategies. These insights, in turn, inform the design of our reinforcement learning-based exploration policy, leveraging global map predictions and other environmental contexts and enabling state-of-the-art performance. Abstract -- Path planning for robotic exploration is challenging, requiring reasoning over unknown spaces and anticipating future observations. Efficient exploration requires selecting budget-constrained paths that maximize information gain. Despite advances in autonomous exploration, existing algorithms still fall short of human performance, particularly in structured environments where predictive cues exist but are underutilized. Guided by insights from our user study, we introduce MapExRL, which improves robot exploration efficiency in structured indoor environments by enabling longer-horizon planning through reinforcement learning (RL) and global map predictions. Our framework generates global map predictions from the observed map, which our policy utilizes, along with the prediction uncertainty, estimated sensor coverage, frontier distance, and remaining distance budget, to assess the strategic long-term value of frontiers. By leveraging multiple frontier scoring methods and additional context, our policy makes more informed decisions at each stage of the exploration. We evaluate our framework on a real-world indoor map dataset, achieving up to an 18.8% improvement over the strongest state-of-the-art baseline, with even greater gains compared to conventional frontier-based algorithms. This work involved human subjects or animals in its research.

-- Interactive navigation is crucial in scenarios where proactively interacting with objects can yield shorter paths, thus significantly improving traversal efficiency. Existing methods primarily focus on using the robot body to relocate large obstacles (which could be comparable to the size of a robot). However, they prove ineffective in narrow or constrained spaces where the robot's dimensions restrict its manipulation capabilities. This paper introduces a novel interactive navigation framework for legged manipulators, featuring an active arm-pushing mechanism that enables the robot to reposition movable obstacles in space-constrained environments. T o this end, we develop a reinforcement learning-based arm-pushing controller with a two-stage reward strategy for large-object manipulation. Specifically, this strategy first directs the manipulator to a designated pushing zone to achieve a kinematically feasible contact configuration. Then, the end effector is guided to maintain its position at appropriate contact points for stable object displacement while preventing toppling. The simulations validate the robustness of the arm-pushing controller, showing that the two-stage reward strategy improves policy convergence and long-term performance. Real-world experiments further demonstrate the effectiveness of the proposed navigation framework, which achieves shorter paths and reduced traversal time.

On-policy reinforcement learning algorithms use the most recently learned policy to interact with the environment and update it using the latest gathered trajectories, making them well-suited for adapting to non-stationary environments where dynamics change over time. However, previous studies show that they struggle to maintain plasticity$\unicode{x2013}$the ability of neural networks to adjust their synaptic connections$\unicode{x2013}$with overfitting identified as the primary cause. To address this, we present the first application of evidential learning in an on-policy reinforcement learning setting: $\textit{Evidential Proximal Policy Optimization (EPPO)}$. EPPO incorporates all sources of error in the critic network's approximation$\unicode{x2013}$i.e., the baseline function in advantage calculation$\unicode{x2013}$by modeling the epistemic and aleatoric uncertainty contributions to the approximation's total variance. We achieve this by using an evidential neural network, which serves as a regularizer to prevent overfitting. The resulting probabilistic interpretation of the advantage function enables optimistic exploration, thus maintaining the plasticity. Through experiments on non-stationary continuous control tasks, where the environment dynamics change at regular intervals, we demonstrate that EPPO outperforms state-of-the-art on-policy reinforcement learning variants in both task-specific and overall return.

Although multi-agent reinforcement learning (MARL) has shown its success across diverse domains, extending its application to large-scale real-world systems still faces significant challenges. Primarily, the high complexity of real-world environments exacerbates the credit assignment problem, substantially reducing training efficiency. Moreover, the variability of agent populations in large-scale scenarios necessitates scalable decision-making mechanisms. To address these challenges, we propose a novel framework: Sequential rollout with Sequential value estimation (SrSv). This framework aims to capture agent interdependence and provide a scalable solution for cooperative MARL. Specifically, SrSv leverages the autoregressive property of the Transformer model to handle varying populations through sequential action rollout. Furthermore, to capture the interdependence of policy distributions and value functions among multiple agents, we introduce an innovative sequential value estimation methodology and integrates the value approximation into an attention-based sequential model. We evaluate SrSv on three benchmarks: Multi-Agent MuJoCo, StarCraft Multi-Agent Challenge, and DubinsCars. Experimental results demonstrate that SrSv significantly outperforms baseline methods in terms of training efficiency without compromising convergence performance. Moreover, when implemented in a large-scale DubinsCar system with 1,024 agents, our framework surpasses existing benchmarks, highlighting the excellent scalability of SrSv.

In the context of multi-agent reinforcement learning, generalization is a challenge to solve various tasks that may require different joint policies or coordination without relying on policies specialized for each task. We refer to this type of problem as a multi-task, and we train agents to be versatile in this multi-task setting through a single training process. To address this challenge, we introduce TRajectory-class-Aware Multi-Agent reinforcement learning (TRAMA). In TRAMA, agents recognize a task type by identifying the class of trajectories they are experiencing through partial observations, and the agents use this trajectory awareness or prediction as additional information for action policy. To this end, we introduce three primary objectives in TRAMA: (a) constructing a quantized latent space to generate trajectory embeddings that reflect key similarities among them; (b) conducting trajectory clustering using these trajectory embeddings; and (c) building a trajectory-class-aware policy. Specifically for (c), we introduce a trajectory-class predictor that performs agent-wise predictions on the trajectory class; and we design a trajectory-class representation model for each trajectory class. Each agent takes actions based on this trajectory-class representation along with its partial observation for task-aware execution. The proposed method is evaluated on various tasks, including multi-task problems built upon StarCraft II. Empirical results show further performance improvements over state-of-the-art baselines.

Recent works have successfully demonstrated that sparse deep reinforcement learning agents can be competitive against their dense counterparts. This opens up opportunities for reinforcement learning applications in fields where inference time and memory requirements are cost-sensitive or limited by hardware. Until now, dense-to-sparse methods have relied on hand-designed sparsity schedules that are not synchronized with the agent's learning pace. Crucially, the final sparsity level is chosen as a hyperparameter, which requires careful tuning as setting it too high might lead to poor performances. In this work, we address these shortcomings by crafting a dense-to-sparse algorithm that we name Eau De $Q$-Network (EauDeQN). To increase sparsity at the agent's learning pace, we consider multiple online networks with different sparsity levels, where each online network is trained from a shared target network. At each target update, the online network with the smallest loss is chosen as the next target network, while the other networks are replaced by a pruned version of the chosen network. We evaluate the proposed approach on the Atari $2600$ benchmark and the MuJoCo physics simulator, showing that EauDeQN reaches high sparsity levels while keeping performances high.