Queuing network control is essential for managing congestion in job-processing systems such as service systems, communication networks, and manufacturing processes. Despite growing interest in applying reinforcement learning (RL) techniques, queueing network control poses distinct challenges, including high stochasticity, large state and action spaces, and lack of stability. To tackle these challenges, we propose a scalable framework for policy optimization based on differentiable discrete event simulation. Our main insight is that by implementing a well-designed smoothing technique for discrete event dynamics, we can compute pathwise policy gradients for large-scale queueing networks using auto-differentiation software (e.g., Tensorflow, PyTorch) and GPU parallelization. Through extensive empirical experiments, we observe that our policy gradient estimators are several orders of magnitude more accurate than typical REINFORCE-based estimators. In addition, We propose a new policy architecture, which drastically improves stability while maintaining the flexibility of neural-network policies. In a wide variety of scheduling and admission control tasks, we demonstrate that training control policies with pathwise gradients leads to a 50-1000x improvement in sample efficiency over state-of-the-art RL methods. Unlike prior tailored approaches to queueing, our methods can flexibly handle realistic scenarios, including systems operating in non-stationary environments and those with non-exponential interarrival/service times.

In the field of autonomous driving, developing safe and trustworthy autonomous driving policies remains a significant challenge. Recently, Reinforcement Learning with Human Feedback (RLHF) has attracted substantial attention due to its potential to enhance training safety and sampling efficiency. Nevertheless, existing RLHF-enabled methods often falter when faced with imperfect human demonstrations, potentially leading to training oscillations or even worse performance than rule-based approaches. Inspired by the human learning process, we propose Physics-enhanced Reinforcement Learning with Human Feedback (PE-RLHF). This novel framework synergistically integrates human feedback (e.g., human intervention and demonstration) and physics knowledge (e.g., traffic flow model) into the training loop of reinforcement learning. The key advantage of PE-RLHF is its guarantee that the learned policy will perform at least as well as the given physics-based policy, even when human feedback quality deteriorates, thus ensuring trustworthy safety improvements. PE-RLHF introduces a Physics-enhanced Human-AI (PE-HAI) collaborative paradigm for dynamic action selection between human and physics-based actions, employs a reward-free approach with a proxy value function to capture human preferences, and incorporates a minimal intervention mechanism to reduce the cognitive load on human mentors. Extensive experiments across diverse driving scenarios demonstrate that PE-RLHF significantly outperforms traditional methods, achieving state-of-the-art (SOTA) performance in safety, efficiency, and generalizability, even with varying quality of human feedback. The philosophy behind PE-RLHF not only advances autonomous driving technology but can also offer valuable insights for other safety-critical domains. Demo video and code are available at: \https://zilin-huang.github.io/PE-RLHF-website/

Performing long-term experimentation or large-scale data collection for machine learning in the field of soft robotics is challenging, due to the hardware robustness and experimental flexibility required. In this work, we propose a modular parallel robotic manipulation platform suitable for such large-scale data collection and compatible with various soft-robotic fabrication methods. Considering the computational and theoretical difficulty of replicating the high-fidelity, faster-than-real-time simulations that enable large-scale data collection in rigid robotic systems, a robust soft-robotic hardware platform becomes a high priority development task for the field. The platform's modules consist of a pair of off-the-shelf electrical motors which actuate a customizable finger consisting of a compliant parallel structure. The parallel mechanism of the finger can be as simple as a single 3D-printed urethane or molded silicone bulk structure, due to the motors being able to fully actuate a passive structure. This design flexibility allows experimentation with soft mechanism varied geometries, bulk properties and surface properties. Additionally, while the parallel mechanism does not require separate electronics or additional parts, these can be included, and it can be constructed using multi-functional soft materials to study compatible soft sensors and actuators in the learning process. In this work, we validate the platform's ability to be used for policy gradient reinforcement learning directly on hardware in a benchmark 2D manipulation task. We additionally demonstrate compatibility with multiple fingers and characterize the design constraints for compatible extensions.

We propose an agent architecture that automates parts of the common reinforcement learning experiment workflow, to enable automated mastery of control domains for embodied agents. To do so, it leverages a VLM to perform some of the capabilities normally required of a human experimenter, including the monitoring and analysis of experiment progress, the proposition of new tasks based on past successes and failures of the agent, decomposing tasks into a sequence of subtasks (skills), and retrieval of the skill to execute - enabling our system to build automated curricula for learning. We believe this is one of the first proposals for a system that leverages a VLM throughout the full experiment cycle of reinforcement learning. We provide a first prototype of this system, and examine the feasibility of current models and techniques for the desired level of automation. For this, we use a standard Gemini model, without additional fine-tuning, to provide a curriculum of skills to a language-conditioned Actor-Critic algorithm, in order to steer data collection so as to aid learning new skills. Data collected in this way is shown to be useful for learning and iteratively improving control policies in a robotics domain. Additional examination of the ability of the system to build a growing library of skills, and to judge the progress of the training of those skills, also shows promising results, suggesting that the proposed architecture provides a potential recipe for fully automated mastery of tasks and domains for embodied agents.

This document contains an educational introduction to the problem of sparsifying parametric models with L0 regularization. We utilize this approach together with dictionary learning to learn sparse polynomial policies for deep reinforcement learning to control parametric partial differential equations. The code and a tutorial are provided here: https://github.com/nicob15/Sparsifying-Parametric-Models-with-L0.

This paper studies asynchronous stochastic approximation (SA) algorithms and their application to reinforcement learning in semi-Markov decision processes (SMDPs) with an average-reward criterion. We first extend Borkar and Meyn's stability proof method to accommodate more general noise conditions, leading to broader convergence guarantees for asynchronous SA algorithms. Leveraging these results, we establish the convergence of an asynchronous SA analogue of Schweitzer's classical relative value iteration algorithm, RVI Q-learning, for finite-space, weakly communicating SMDPs. Furthermore, to fully utilize the SA results in this application, we introduce new monotonicity conditions for estimating the optimal reward rate in RVI Q-learning. These conditions substantially expand the previously considered algorithmic framework, and we address them with novel proof arguments in the stability and convergence analysis of RVI Q-learning.

This book uses the modern theory of artificial intelligence (AI) to understand human suffering or mental pain. Both humans and sophisticated AI agents process information about the world in order to achieve goals and obtain rewards, which is why AI can be used as a model of the human brain and mind. This book intends to make the theory accessible to a relatively general audience, requiring only some relevant scientific background. The book starts with the assumption that suffering is mainly caused by frustration. Frustration means the failure of an agent (whether AI or human) to achieve a goal or a reward it wanted or expected. Frustration is inevitable because of the overwhelming complexity of the world, limited computational resources, and scarcity of good data. In particular, such limitations imply that an agent acting in the real world must cope with uncontrollability, unpredictability, and uncertainty, which all lead to frustration. Fundamental in such modelling is the idea of learning, or adaptation to the environment. While AI uses machine learning, humans and animals adapt by a combination of evolutionary mechanisms and ordinary learning. Even frustration is fundamentally an error signal that the system uses for learning. This book explores various aspects and limitations of learning algorithms and their implications regarding suffering. At the end of the book, the computational theory is used to derive various interventions or training methods that will reduce suffering in humans. The amount of frustration is expressed by a simple equation which indicates how it can be reduced. The ensuing interventions are very similar to those proposed by Buddhist and Stoic philosophy, and include mindfulness meditation. Therefore, this book can be interpreted as an exposition of a computational theory justifying why such philosophies and meditation reduce human suffering.

Reinforcement learning agents can achieve superhuman performance in static tasks but are costly to train and fragile to task changes. This limits their deployment in real-world scenarios where training experience is expensive or the context changes through factors like sensor degradation, environmental processes or changing mission priorities. Lifelong reinforcement learning aims to improve sample efficiency and adaptability by studying how agents perform in evolving problems. The difficulty that these changes pose to an agent is rarely measured directly, however. Agent performances can be compared across a change, but this is often prohibitively expensive. We propose Change-Induced Regret Proxy (CHIRP) metrics, a class of metrics for approximating a change's difficulty while avoiding the high costs of using trained agents. A relationship between a CHIRP metric and agent performance is identified in two environments, a simple grid world and MetaWorld's suite of robotic arm tasks. We demonstrate two uses for these metrics: for learning, an agent that clusters MDPs based on a CHIRP metric achieves $17\%$ higher average returns than three existing agents in a sequence of MetaWorld tasks. We also show how a CHIRP can be calibrated to compare the difficulty of changes across distinctly different environments.

Autonomous indoor navigation of UAVs presents numerous challenges, primarily due to the limited precision of GPS in enclosed environments. Additionally, UAVs' limited capacity to carry heavy or power-intensive sensors, such as overheight packages, exacerbates the difficulty of achieving autonomous navigation indoors. This paper introduces an advanced system in which a drone autonomously navigates indoor spaces to locate a specific target, such as an unknown Amazon package, using only a single camera. Employing a deep learning approach, a deep reinforcement adaptive learning algorithm is trained to develop a control strategy that emulates the decision-making process of an expert pilot. We demonstrate the efficacy of our system through real-time simulations conducted in various indoor settings. We apply multiple visualization techniques to gain deeper insights into our trained network. Furthermore, we extend our approach to include an adaptive control algorithm for coordinating multiple drones to lift an object in an indoor environment collaboratively. Integrating our DRAL algorithm enables multiple UAVs to learn optimal control strategies that adapt to dynamic conditions and uncertainties. This innovation enhances the robustness and flexibility of indoor navigation and opens new possibilities for complex multi-drone operations in confined spaces. The proposed framework highlights significant advancements in adaptive control and deep reinforcement learning, offering robust solutions for complex multi-agent systems in real-world applications.

Reinforcement Learning (RL) is highly dependent on the meticulous design of the reward function. However, accurately assigning rewards to each state-action pair in Long-Term RL (LTRL) challenges is formidable. Consequently, RL agents are predominantly trained with expert guidance. Drawing on the principles of ordinal utility theory from economics, we propose a novel reward estimation algorithm: ELO-Rating based RL (ERRL). This approach is distinguished by two main features. Firstly, it leverages expert preferences over trajectories instead of cardinal rewards (utilities) to compute the ELO rating of each trajectory as its reward. Secondly, a new reward redistribution algorithm is introduced to mitigate training volatility in the absence of a fixed anchor reward. Our method demonstrates superior performance over several leading baselines in long-term scenarios (extending up to 5000 steps), where conventional RL algorithms falter. Furthermore, we conduct a thorough analysis of how expert preferences affect the outcomes.