Partial monitoring is an expressive framework for sequential decision-making with an abundance of applications, including graph-structured and dueling bandits, dynamic pricing and transductive feedback models. We survey and extend recent results on the linear formulation of partial monitoring that naturally generalizes the standard linear bandit setting. The main result is that a single algorithm, information-directed sampling (IDS), is (nearly) worst-case rate optimal in all finite-action games. We present a simple and unified analysis of stochastic partial monitoring, and further extend the model to the contextual and kernelized setting.

Ensuring safety of nonlinear systems under model uncertainty and external disturbances is crucial, especially for real-world control tasks. Predictive methods such as robust model predictive control (RMPC) require solving nonconvex optimization problems online, which leads to high computational burden and poor scalability. Reinforcement learning (RL) works well with complex systems, but pays the price of losing rigorous safety guarantee. This paper presents a theoretical framework that bridges the advantages of both RMPC and RL to synthesize safety filters for nonlinear systems with state- and action-dependent uncertainty. We decompose the robust invariant set (RIS) into two parts: a target set that aligns with terminal region design of RMPC, and a reach-avoid set that accounts for the rest of RIS. We propose a policy iteration approach for robust reach-avoid problems and establish its monotone convergence. This method sets the stage for an adversarial actor-critic deep RL algorithm, which simultaneously synthesizes a reach-avoid policy network, a disturbance policy network, and a reach-avoid value network. The learned reach-avoid policy network is utilized to generate nominal trajectories for online verification, which filters potentially unsafe actions that may drive the system into unsafe regions when worst-case disturbances are applied. We formulate a second-order cone programming (SOCP) approach for online verification using system level synthesis, which optimizes for the worst-case reach-avoid value of any possible trajectories. The proposed safety filter requires much lower computational complexity than RMPC and still enjoys persistent robust safety guarantee. The effectiveness of our method is illustrated through a numerical example.

We study fair multi-objective reinforcement learning in which an agent must learn a policy that simultaneously achieves high reward on multiple dimensions of a vector-valued reward. Motivated by the fair resource allocation literature, we model this as an expected welfare maximization problem, for some nonlinear fair welfare function of the vector of long-term cumulative rewards. One canonical example of such a function is the Nash Social Welfare, or geometric mean, the log transform of which is also known as the Proportional Fairness objective. We show that even approximately optimal optimization of the expected Nash Social Welfare is computationally intractable even in the tabular case. Nevertheless, we provide a novel adaptation of Q-learning that combines nonlinear scalarized learning updates and non-stationary action selection to learn effective policies for optimizing nonlinear welfare functions. We show that our algorithm is provably convergent, and we demonstrate experimentally that our approach outperforms techniques based on linear scalarization, mixtures of optimal linear scalarizations, or stationary action selection for the Nash Social Welfare Objective.

Federated Learning (FL) has emerged as a promising Machine Learning paradigm, enabling multiple users to collaboratively train a shared model while preserving their local data. To minimize computing and communication costs associated with parameter transfer, it is common practice in FL to select a subset of clients in each training round. This selection must consider both system and static heterogeneity. Therefore, we propose FLASH-RL, a framework that utilizes Double Deep QLearning (DDQL) to address both system and static heterogeneity in FL. FLASH-RL introduces a new reputation-based utility function to evaluate client contributions based on their current and past performances. Additionally, an adapted DDQL algorithm is proposed to expedite the learning process. Experimental results on MNIST and CIFAR-10 datasets have shown FLASH-RL's effectiveness in achieving a balanced trade-off between model performance and end-to-end latency against existing solutions. Indeed, FLASH-RL reduces latency by up to 24.83% compared to FedAVG and 24.67% compared to FAVOR. It also reduces the training rounds by up to 60.44% compared to FedAVG and +76% compared to FAVOR. In fall detection using the MobiAct dataset, FLASH-RL outperforms FedAVG by up to 2.82% in model's performance and reduces latency by up to 34.75%. Additionally, FLASH-RL achieves the target performance faster, with up to a 45.32% reduction in training rounds compared to FedAVG.

In recent times, reinforcement learning has produced baffling results when it comes to performing control tasks with highly non-linear systems. The impressive results always outweigh the potential vulnerabilities or uncertainties associated with the agents when deployed in the real-world. While the performance is remarkable compared to the classical control algorithms, the reinforcement learning-based methods suffer from two flaws, robustness and interpretability, which are vital for contemporary real-world applications. The paper attempts to alleviate such problems with reinforcement learning and proposes the concept of model-assisted reinforcement learning to induce a notion of conservativeness in the agents. The control task considered for the experiment involves navigating a CrazyFlie quadrotor. The paper also describes a way of reformulating the task to have the flexibility of tuning the level of conservativeness via multi-objective reinforcement learning. The results include a comparison of the vanilla reinforcement learning approaches and the proposed approach. The metrics are evaluated by systematically injecting disturbances to classify the inherent robustness and conservativeness of the agents. More concrete arguments are made by computing and comparing the backward reachability tubes of the RL policies by solving the Hamilton-Jacobi-Bellman partial differential equation (HJ PDE).

Federated Learning (FL) is a rapidly growing field in machine learning that allows data to be trained across multiple decentralized devices. The selection of clients to participate in the training process is a critical factor for the performance of the overall system. In this survey, we provide a comprehensive overview of the state-of-the-art client selection techniques in FL, including their strengths and limitations, as well as the challenges and open issues that need to be addressed. We cover conventional selection techniques such as random selection where all or partial random of clients is used for the trained. We also cover performance-aware selections and as well as resource-aware selections for resource-constrained networks and heterogeneous networks. We also discuss the usage of client selection in model security enhancement. Lastly, we discuss open issues and challenges related to clients selection in dynamic constrained, and heterogeneous networks.

The goal of program synthesis, or code generation, is to generate executable code based on given descriptions. Recently, there has been an increasing number of studies employing reinforcement learning (RL) to improve the performance of large language models (LLMs) for code. However, current representative works either rely solely on offline frameworks, limiting the exploration of new sample spaces, or fall short in the utilization of unit test signals, not accounting for specific error locations within the code. To address these issues, we propose RLTF, i.e., Reinforcement Learning from Unit Test Feedback, a novel online RL framework with unit test feedback of multi-granularity for refining code LLMs. Our approach generates data in real-time during training and simultaneously utilizes fine-grained feedback signals to guide the model towards producing higher-quality code. Extensive experiments show that RLTF achieves state-of-the-art performance on the APPS and the MBPP benchmarks. Our code is available at: https://github.com/Zyq-scut/RLTF.

This paper presents an innovative method for humanoid robots to acquire a comprehensive set of motor skills through reinforcement learning. The approach utilizes an achievement-triggered multi-path reward function rooted in developmental robotics principles, facilitating the robot to learn gross motor skills typically mastered by human infants within a single training phase. The proposed method outperforms standard reinforcement learning techniques in success rates and learning speed within a simulation environment. By leveraging the principles of self-discovery and exploration integral to infant learning, this method holds the potential to significantly advance humanoid robot motor skill acquisition.

Deep Reinforcement Learning (DRL) has shown promise in various networking environments. However, these environments present several fundamental challenges for standard DRL techniques. They are difficult to explore and exhibit high levels of noise and uncertainty. Although these challenges complicate the training process, we find that in practice we can substantially mitigate their effects and even achieve state-of-the-art real-world performance by addressing a factor that has been previously overlooked: the skewed input trace distribution in DRL training datasets. We introduce a generalized framework, Plume, to automatically identify and balance the skew using a three-stage process. First, we identify the critical features that determine the behavior of the traces. Second, we classify the traces into clusters. Finally, we prioritize the salient clusters to improve the overall performance of the controller. Plume seamlessly works across DRL algorithms, without requiring any changes to the DRL workflow. We evaluated Plume on three networking environments, including Adaptive Bitrate Streaming, Congestion Control, and Load Balancing. Plume offers superior performance in both simulation and real-world settings, across different controllers and DRL algorithms. For example, our novel ABR controller, Gelato trained with Plume consistently outperforms prior state-of-the-art controllers on the live streaming platform Puffer for over a year. It is the first controller on the platform to deliver statistically significant improvements in both video quality and stalling, decreasing stalls by as much as 75%.

Robotic skill learning has been increasingly studied but the demonstration collections are more challenging compared to collecting images/videos in computer vision and texts in natural language processing. This paper presents a skill learning paradigm by using intuitive teleoperation devices to generate high-quality human demonstrations efficiently for robotic skill learning in a data-driven manner. By using a reliable teleoperation interface, the da Vinci Research Kit (dVRK) master, a system called dVRK-Simulator-for-Demonstration (dS4D) is proposed in this paper. Various manipulation tasks show the system's effectiveness and advantages in efficiency compared to other interfaces. Using the collected data for policy learning has been investigated, which verifies the initial feasibility. We believe the proposed paradigm can facilitate robot learning driven by high-quality demonstrations and efficiency while generating them.