Reinforcement learning is a powerful approach for training an optimal policy to solve complex problems in a given system. This project aims to demonstrate the application of reinforcement learning in stochastic process environments with missing information, using Flappy Bird and a newly developed stock trading environment as case studies. We evaluate various structures of Deep Q-learning networks and identify the most suitable variant for the stochastic process environment. Additionally, we discuss the current challenges and propose potential improvements for further work in environment-building and reinforcement learning techniques.

Ultrasound (US) is one of the most widely used modalities for clinical intervention and diagnosis due to the merits of providing non-invasive, radiation-free, and real-time images. However, free-hand US examinations are highly operator-dependent. Robotic US System (RUSS) aims at overcoming this shortcoming by offering reproducibility, while also aiming at improving dexterity, and intelligent anatomy and disease-aware imaging. In addition to enhancing diagnostic outcomes, RUSS also holds the potential to provide medical interventions for populations suffering from the shortage of experienced sonographers. In this paper, we categorize RUSS as teleoperated or autonomous. Regarding teleoperated RUSS, we summarize their technical developments, and clinical evaluations, respectively. This survey then focuses on the review of recent work on autonomous robotic US imaging. We demonstrate that machine learning and artificial intelligence present the key techniques, which enable intelligent patient and process-specific, motion and deformation-aware robotic image acquisition. We also show that the research on artificial intelligence for autonomous RUSS has directed the research community toward understanding and modeling expert sonographers' semantic reasoning and action. Here, we call this process, the recovery of the "language of sonography". This side result of research on autonomous robotic US acquisitions could be considered as valuable and essential as the progress made in the robotic US examination itself. This article will provide both engineers and clinicians with a comprehensive understanding of RUSS by surveying underlying techniques.

There is a growing interest in using reinforcement learning (RL) to personalize sequences of treatments in digital health to support users in adopting healthier behaviors. Such sequential decision-making problems involve decisions about when to treat and how to treat based on the user's context (e.g., prior activity level, location, etc.). Online RL is a promising data-driven approach for this problem as it learns based on each user's historical responses and uses that knowledge to personalize these decisions. However, to decide whether the RL algorithm should be included in an ``optimized'' intervention for real-world deployment, we must assess the data evidence indicating that the RL algorithm is actually personalizing the treatments to its users. Due to the stochasticity in the RL algorithm, one may get a false impression that it is learning in certain states and using this learning to provide specific treatments. We use a working definition of personalization and introduce a resampling-based methodology for investigating whether the personalization exhibited by the RL algorithm is an artifact of the RL algorithm stochasticity. We illustrate our methodology with a case study by analyzing the data from a physical activity clinical trial called HeartSteps, which included the use of an online RL algorithm. We demonstrate how our approach enhances data-driven truth-in-advertising of algorithm personalization both across all users as well as within specific users in the study.

Earth imaging satellites are a crucial part of our everyday lives that enable global tracking of industrial activities. Use cases span many applications, from weather forecasting to digital maps, carbon footprint tracking, and vegetation monitoring. However, there are limitations; satellites are difficult to manufacture, expensive to maintain, and tricky to launch into orbit. Therefore, satellites must be employed efficiently. This poses a challenge known as the satellite mission planning problem, which could be computationally prohibitive to solve on large scales. However, close-to-optimal algorithms, such as greedy reinforcement learning and optimization algorithms, can often provide satisfactory resolutions. This paper introduces a set of quantum algorithms to solve the mission planning problem and demonstrate an advantage over the classical algorithms implemented thus far. The problem is formulated as maximizing the number of high-priority tasks completed on real datasets containing thousands of tasks and multiple satellites. This work demonstrates that through solution-chaining and clustering, optimization and machine learning algorithms offer the greatest potential for optimal solutions. This paper notably illustrates that a hybridized quantum-enhanced reinforcement learning agent can achieve a completion percentage of 98.5% over high-priority tasks, significantly improving over the baseline greedy methods with a completion rate of 75.8%. The results presented in this work pave the way to quantum-enabled solutions in the space industry and, more generally, future mission planning problems across industries.

In this paper, we present a general learning framework for controlling a quadruped robot that can mimic the behavior of real animals and traverse challenging terrains. Our method consists of two steps: an imitation learning step to learn from motions of real animals, and a terrain adaptation step to enable generalization to unseen terrains. We capture motions from a Labrador on various terrains to facilitate terrain adaptive locomotion. Our experiments demonstrate that our policy can traverse various terrains and produce a natural-looking behavior. We deployed our method on the real quadruped robot Max via zero-shot simulation-to-reality transfer, achieving a speed of 1.1 m/s on stairs climbing.

Non-stationarity is a fundamental challenge in multi-agent reinforcement learning (MARL), where agents update their behaviour as they learn. Many theoretical advances in MARL avoid the challenge of non-stationarity by coordinating the policy updates of agents in various ways, including synchronizing times at which agents are allowed to revise their policies. Synchronization enables analysis of many MARL algorithms via multi-timescale methods, but such synchrony is infeasible in many decentralized applications. In this paper, we study an asynchronous variant of the decentralized Q-learning algorithm, a recent MARL algorithm for stochastic games. We provide sufficient conditions under which the asynchronous algorithm drives play to equilibrium with high probability. Our solution utilizes constant learning rates in the Q-factor update, which we show to be critical for relaxing the synchrony assumptions of earlier work. Our analysis also applies to asynchronous generalizations of a number of other algorithms from the regret testing tradition, whose performance is analyzed by multi-timescale methods that study Markov chains obtained via policy update dynamics. This work extends the applicability of the decentralized Q-learning algorithm and its relatives to settings in which parameters are selected in an independent manner, and tames non-stationarity without imposing the coordination assumptions of prior work.

Interactive reinforcement learning has shown promise in learning complex robotic tasks. However, the process can be human-intensive due to the requirement of a large amount of interactive feedback. This paper presents a new method that uses scores provided by humans instead of pairwise preferences to improve the feedback efficiency of interactive reinforcement learning. Our key insight is that scores can yield significantly more data than pairwise preferences. Specifically, we require a teacher to interactively score the full trajectories of an agent to train a behavioral policy in a sparse reward environment. To avoid unstable scores given by humans negatively impacting the training process, we propose an adaptive learning scheme. This enables the learning paradigm to be insensitive to imperfect or unreliable scores. We extensively evaluate our method for robotic locomotion and manipulation tasks. The results show that the proposed method can efficiently learn near-optimal policies by adaptive learning from scores while requiring less feedback compared to pairwise preference learning methods. The source codes are publicly available at https://github.com/SSKKai/Interactive-Scoring-IRL.

Simulation is essential to reinforcement learning (RL) before implementation in the real world, especially for safety-critical applications like robot manipulation. Conventionally, RL agents are sensitive to the discrepancies between the simulation and the real world, known as the sim-to-real gap. The application of domain randomization, a technique used to fill this gap, is limited to the imposition of heuristic-randomized models. {We investigate the properties of intrinsic stochasticity of real-time simulation (RT-IS) of off-the-shelf simulation software and its potential to improve RL performance. This improvement includes a higher tolerance to noise and model imprecision and superiority to conventional domain randomization in terms of ease of use and automation. Firstly, we conduct analytical studies to measure the correlation of RT-IS with the utilization of computer hardware and validate its comparability with the natural stochasticity of a physical robot. Then, we exploit the RT-IS feature in the training of an RL agent. The simulation and physical experiment results verify the feasibility and applicability of RT-IS to robust agent training for robot manipulation tasks. The RT-IS-powered RL agent outperforms conventional agents on robots with modeling uncertainties. RT-IS requires less heuristic randomization, is not task-dependent, and achieves better generalizability than the conventional domain-randomization-powered agents. Our findings provide a new perspective on the sim-to-real problem in practical applications like robot manipulation tasks.

We study a new paradigm for sequential decision making, called offline policy learning from observations (PLfO). Offline PLfO aims to learn policies using datasets with substandard qualities: 1) only a subset of trajectories is labeled with rewards, 2) labeled trajectories may not contain actions, 3) labeled trajectories may not be of high quality, and 4) the data may not have full coverage. Such imperfection is common in real-world learning scenarios, and offline PLfO encompasses many existing offline learning setups, including offline imitation learning (IL), offline IL from observations (ILfO), and offline reinforcement learning (RL). In this work, we present a generic approach to offline PLfO, called $\textbf{M}$odality-agnostic $\textbf{A}$dversarial $\textbf{H}$ypothesis $\textbf{A}$daptation for $\textbf{L}$earning from $\textbf{O}$bservations (MAHALO). Built upon the pessimism concept in offline RL, MAHALO optimizes the policy using a performance lower bound that accounts for uncertainty due to the dataset's insufficient coverage. We implement this idea by adversarially training data-consistent critic and reward functions, which forces the learned policy to be robust to data deficiency. We show that MAHALO consistently outperforms or matches specialized algorithms across a variety of offline PLfO tasks in theory and experiments. Our code is available at https://github.com/AnqiLi/mahalo.

Due to the emergence of AI systems that interact with the physical environment, there is an increased interest in incorporating physical reasoning capabilities into those AI systems. But is it enough to only have physical reasoning capabilities to operate in a real physical environment? In the real world, we constantly face novel situations we have not encountered before. As humans, we are competent at successfully adapting to those situations. Similarly, an agent needs to have the ability to function under the impact of novelties in order to properly operate in an open-world physical environment. To facilitate the development of such AI systems, we propose a new testbed, NovPhy, that requires an agent to reason about physical scenarios in the presence of novelties and take actions accordingly. The testbed consists of tasks that require agents to detect and adapt to novelties in physical scenarios. To create tasks in the testbed, we develop eight novelties representing a diverse novelty space and apply them to five commonly encountered scenarios in a physical environment. According to our testbed design, we evaluate two capabilities of an agent: the performance on a novelty when it is applied to different physical scenarios and the performance on a physical scenario when different novelties are applied to it. We conduct a thorough evaluation with human players, learning agents, and heuristic agents. Our evaluation shows that humans' performance is far beyond the agents' performance. Some agents, even with good normal task performance, perform significantly worse when there is a novelty, and the agents that can adapt to novelties typically adapt slower than humans. We promote the development of intelligent agents capable of performing at the human level or above when operating in open-world physical environments. Testbed website: https://github.com/phy-q/novphy