In this paper, we present AR3n (pronounced as Aaron), an assist-as-needed (AAN) controller that utilizes reinforcement learning to supply adaptive assistance during a robot assisted handwriting rehabilitation task. Unlike previous AAN controllers, our method does not rely on patient specific controller parameters or physical models. We propose the use of a virtual patient model to generalize AR3n across multiple subjects. The system modulates robotic assistance in realtime based on a subject's tracking error, while minimizing the amount of robotic assistance. The controller is experimentally validated through a set of simulations and human subject experiments. Finally, a comparative study with a traditional rule-based controller is conducted to analyze differences in assistance mechanisms of the two controllers.

In the last course of our specialization, Overview of Advanced Methods of Reinforcement Learning in Finance, we will take a deeper look into topics discussed in our third course, Reinforcement Learning in Finance. In particular, we will talk about links between Reinforcement Learning, option pricing and physics, implications of Inverse Reinforcement Learning for modeling market impact and price dynamics, and perception-action cycles in Reinforcement Learning. After taking this course, students will be able to - explain fundamental concepts of finance such as market equilibrium, no arbitrage, predictability, - discuss market modeling, - Apply the methods of Reinforcement Learning to high-frequency trading, credit risk peer-to-peer lending, and cryptocurrencies trading.

In the field of motion simulation, the level of immersion strongly depends on the motion cueing algorithm (MCA), as it transfers the reference motion of the simulated vehicle to a motion of the motion simulation platform (MSP). The challenge for the MCA is to reproduce the motion perception of a real vehicle driver as accurately as possible without exceeding the limits of the workspace of the MSP in order to provide a realistic virtual driving experience. In case of a large discrepancy between the perceived motion signals and the optical cues, motion sickness may occur with the typical symptoms of nausea, dizziness, headache and fatigue. Existing approaches either produce non-optimal results, e.g., due to filtering, linearization, or simplifications, or the required computational time exceeds the real-time requirements of a closed-loop application. In this work a new solution is presented, where not a human designer specifies the principles of the MCA but an artificial intelligence (AI) learns the optimal motion by trial and error in an interaction with the MSP. To achieve this, deep reinforcement learning (RL) is applied, where an agent interacts with an environment formulated as a Markov decision process~(MDP). This allows the agent to directly control a simulated MSP to obtain feedback on its performance in terms of platform workspace usage and the motion acting on the simulator user. The RL algorithm used is proximal policy optimization (PPO), where the value function and the policy corresponding to the control strategy are learned and both are mapped in artificial neural networks (ANN). This approach is implemented in Python and the functionality is demonstrated by the practical example of pre-recorded lateral maneuvers. The subsequent validation on a standardized double lane change shows that the RL algorithm is able to learn the control strategy and improve the quality of...

Pre-training with offline data and online fine-tuning using reinforcement learning is a promising strategy for learning control policies by leveraging the best of both worlds in terms of sample efficiency and performance. One natural approach is to initialize the policy for online learning with the one trained offline. In this work, we introduce a policy expansion scheme for this task. After learning the offline policy, we use it as one candidate policy in a policy set. We then expand the policy set with another policy which will be responsible for further learning. The two policies will be composed in an adaptive manner for interacting with the environment. With this approach, the policy previously learned offline is fully retained during online learning, thus mitigating the potential issues such as destroying the useful behaviors of the offline policy in the initial stage of online learning while allowing the offline policy participate in the exploration naturally in an adaptive manner. Moreover, new useful behaviors can potentially be captured by the newly added policy through learning. Experiments are conducted on a number of tasks and the results demonstrate the effectiveness of the proposed approach.

Abstract: The setting of an agent making decisions under uncertainty and under dynamic constraints is common for the fields of optimal control, reinforcement learning, and recently also for online learning. In the online learning setting, the quality of an agent's decision is often quantified by the concept of regret, comparing the performance of the chosen decisions to the best possible ones in hindsight. While regret is a useful performance measure, when dynamical systems are concerned, it is important to also assess the stability of the closed-loop system for a chosen policy. In this work, we show that for linear state feedback policies and linear systems subject to adversarial disturbances, linear regret implies asymptotic stability in both time-varying and time-invariant settings. Conversely, we also show that bounded input bounded state stability and summability of the state transition matrices imply linear regret.

Context: Machine learning (ML) may enable effective automated test generation. Objective: We characterize emerging research, examining testing practices, researcher goals, ML techniques applied, evaluation, and challenges. Methods: We perform a systematic mapping on a sample of 124 publications. Results: ML generates input for system, GUI, unit, performance, and combinatorial testing or improves the performance of existing generation methods. ML is also used to generate test verdicts, property-based, and expected output oracles. Supervised learning - often based on neural networks - and reinforcement learning - often based on Q-learning - are common, and some publications also employ unsupervised or semi-supervised learning. (Semi-/Un-)Supervised approaches are evaluated using both traditional testing metrics and ML-related metrics (e.g., accuracy), while reinforcement learning is often evaluated using testing metrics tied to the reward function. Conclusion: Work-to-date shows great promise, but there are open challenges regarding training data, retraining, scalability, evaluation complexity, ML algorithms employed - and how they are applied - benchmarks, and replicability. Our findings can serve as a roadmap and inspiration for researchers in this field.

Over the last few years, we have not seen any major developments in model-free or model-based learning methods that would make one obsolete relative to the other. In most cases, the used technique is heavily dependent on the use case scenario or other attributes, e.g. the environment. Both approaches have their own advantages, for example, sample efficiency or computational efficiency. However, when combining the two, the advantages of each can be combined and hence achieve better performance. The TD-MPC framework is an example of this approach. On the one hand, a world model in combination with model predictive control is used to get a good initial estimate of the value function. On the other hand, a Q function is used to provide a good long-term estimate. Similar to algorithms like MuZero a latent state representation is used, where only task-relevant information is encoded to reduce the complexity. In this paper, we propose the use of a reconstruction function within the TD-MPC framework, so that the agent can reconstruct the original observation given the internal state representation. This allows our agent to have a more stable learning signal during training and also improves sample efficiency. Our proposed addition of another loss term leads to improved performance on both state- and image-based tasks from the DeepMind-Control suite.

The most relevant problems in discounted reinforcement learning involve estimating the mean of a function under the stationary distribution of a Markov reward process, such as the expected return in policy evaluation, or the policy gradient in policy optimization. In practice, these estimates are produced through a finite-horizon episodic sampling, which neglects the mixing properties of the Markov process. It is mostly unclear how this mismatch between the practical and the ideal setting affects the estimation, and the literature lacks a formal study on the pitfalls of episodic sampling, and how to do it optimally. In this paper, we present a minimax lower bound on the discounted mean estimation problem that explicitly connects the estimation error with the mixing properties of the Markov process and the discount factor. Then, we provide a statistical analysis on a set of notable estimators and the corresponding sampling procedures, which includes the finite-horizon estimators often used in practice. Crucially, we show that estimating the mean by directly sampling from the discounted kernel of the Markov process brings compelling statistical properties w.r.t. the alternative estimators, as it matches the lower bound without requiring a careful tuning of the episode horizon.

Deep Reinforcement Learning (DRL) has shown remarkable success in solving complex tasks across various research fields. However, transferring DRL agents to the real world is still challenging due to the significant discrepancies between simulation and reality. To address this issue, we propose a robust DRL framework that leverages platform-dependent perception modules to extract task-relevant information and train a lane-following and overtaking agent in simulation. This framework facilitates the seamless transfer of the DRL agent to new simulated environments and the real world with minimal effort. We evaluate the performance of the agent in various driving scenarios in both simulation and the real world, and compare it to human players and the PID baseline in simulation. Our proposed framework significantly reduces the gaps between different platforms and the Sim2Real gap, enabling the trained agent to achieve similar performance in both simulation and the real world, driving the vehicle effectively.

We consider the problem of finding the optimal value of n in the n-step temporal difference (TD) learning algorithm. We find the optimal n by resorting to a model-free optimization technique involving a one-simulation simultaneous perturbation stochastic approximation (SPSA) based procedure that we adopt to the discrete optimization setting by using a random projection approach. We prove the convergence of our proposed algorithm, SDPSA, using a differential inclusions approach and show that it finds the optimal value of n in n-step TD. Through experiments, we show that the optimal value of n is achieved with SDPSA for arbitrary initial values. I. INTRODUCTION Reinforcement learning (RL) algorithms are widely used for solving problems of sequential decisionmaking under uncertainty. An RL agent typically makes decisions based on data that it collects through interactions with the environment in order to maximize a certain long-term reward [1], [2].