Optical satellite-to-ground communication (OSGC) has the potential to improve access to fast and affordable Internet in remote regions. Atmospheric turbulence, however, distorts the optical beam, eroding the data rate potential when coupling into single-mode fibers. Traditional adaptive optics (AO) systems use a wavefront sensor to improve fiber coupling. This leads to higher system size, cost and complexity, consumes a fraction of the incident beam and introduces latency, making OSGC for internet service impractical. We propose the use of reinforcement learning (RL) to reduce the latency, size and cost of the system by up to $30-40\%$ by learning a control policy through interactions with a low-cost quadrant photodiode rather than a wavefront phase profiling camera. We develop and share an AO RL environment that provides a standardized platform to develop and evaluate RL based on the Strehl ratio, which is correlated to fiber-coupling performance. Our empirical analysis finds that Proximal Policy Optimization (PPO) outperforms Soft-Actor-Critic and Deep Deterministic Policy Gradient. PPO converges to within $86\%$ of the maximum reward obtained by an idealized Shack-Hartmann sensor after training of 250 episodes, indicating the potential of RL to enable efficient wavefront sensorless OSGC.

Direct Low Earth Orbit satellite-to-handheld links are expected to be part of a new era in satellite communications. Space-Division Multiple Access precoding is a technique that reduces interference among satellite beams, therefore increasing spectral efficiency by allowing cooperating satellites to reuse frequency. Over the past decades, optimal precoding solutions with perfect channel state information have been proposed for several scenarios, whereas robust precoding with only imperfect channel state information has been mostly studied for simplified models. In particular, for Low Earth Orbit satellite applications such simplified models might not be accurate. In this paper, we use the function approximation capabilities of the Soft Actor-Critic deep Reinforcement Learning algorithm to learn robust precoding with no knowledge of the system imperfections.

With the impact of real-time processing being realized in the recent past, the need for efficient implementations of reinforcement learning algorithms has been on the rise. Albeit the numerous advantages of Bellman equations utilized in RL algorithms, they are not without the large search space of design parameters. This research aims to shed light on the design space exploration associated with reinforcement learning parameters, specifically that of Policy Iteration. Given the large computational expenses of fine-tuning the parameters of reinforcement learning algorithms, we propose an auto-tuner-based ordinal regression approach to accelerate the process of exploring these parameters and, in return, accelerate convergence towards an optimal policy. Our approach provides 1.82x peak speedup with an average of 1.48x speedup over the previous state-of-the-art.

State representation learning aims to capture latent factors of an environment. Contrastive methods have performed better than generative models in previous state representation learning research. Although some researchers realize the connections between masked image modeling and contrastive representation learning, the effort is focused on using masks as an augmentation technique to represent the latent generative factors better. Partially observable environments in reinforcement learning have not yet been carefully studied using unsupervised state representation learning methods. In this article, we create an unsupervised state representation learning scheme for partially observable states. We conducted our experiment on a previous Atari 2600 framework designed to evaluate representation learning models. A contrastive method called Spatiotemporal DeepInfomax (ST-DIM) has shown state-of-the-art performance on this benchmark but remains inferior to its supervised counterpart. Our approach improves ST-DIM when the environment is not fully observable and achieves higher F1 scores and accuracy scores than the supervised learning counterpart. The mean accuracy score averaged over categories of our approach is ~66%, compared to ~38% of supervised learning. The mean F1 score is ~64% to ~33%.

Deep Reinforcement Learning is quickly becoming a popular method for training autonomous Unmanned Aerial Vehicles (UAVs). Our work analyzes the effects of measurement uncertainty on the performance of Deep Reinforcement Learning (DRL) based waypoint navigation and obstacle avoidance for UAVs. Measurement uncertainty originates from noise in the sensors used for localization and detecting obstacles. Measurement uncertainty/noise is considered to follow a Gaussian probability distribution with unknown non-zero mean and variance. We evaluate the performance of a DRL agent trained using the Proximal Policy Optimization (PPO) algorithm in an environment with continuous state and action spaces. The environment is randomized with different numbers of obstacles for each simulation episode in the presence of varying degrees of noise, to capture the effects of realistic sensor measurements. Denoising techniques like the low pass filter and Kalman filter improve performance in the presence of unbiased noise. Moreover, we show that artificially injecting noise into the measurements during evaluation actually improves performance in certain scenarios. Extensive training and testing of the DRL agent under various UAV navigation scenarios are performed in the PyBullet physics simulator. To evaluate the practical validity of our method, we port the policy trained in simulation onto a real UAV without any further modifications and verify the results in a real-world environment.

Deep neural networks have been successful in many reinforcement learning settings. However, compared to human learners they are overly data hungry. To build a sample-efficient world model, we apply a transformer to real-world episodes in an autoregressive manner: not only the compact latent states and the taken actions but also the experienced or predicted rewards are fed into the transformer, so that it can attend flexibly to all three modalities at different time steps. The transformer allows our world model to access previous states directly, instead of viewing them through a compressed recurrent state. By utilizing the Transformer-XL architecture, it is able to learn long-term dependencies while staying computationally efficient. Our transformer-based world model (TWM) generates meaningful, new experience, which is used to train a policy that outperforms previous model-free and model-based reinforcement learning algorithms on the Atari 100k benchmark.

Robot assistants for older adults and people with disabilities need to interact with their users in collaborative tasks. The core component of these systems is an interaction manager whose job is to observe and assess the task, and infer the state of the human and their intent to choose the best course of action for the robot. Due to the sparseness of the data in this domain, the policy for such multi-modal systems is often crafted by hand; as the complexity of interactions grows this process is not scalable. In this paper, we propose a reinforcement learning (RL) approach to learn the robot policy. In contrast to the dialog systems, our agent is trained with a simulator developed by using human data and can deal with multiple modalities such as language and physical actions. We conducted a human study to evaluate the performance of the system in the interaction with a user. Our designed system shows promising preliminary results when it is used by a real user.

Reinforcement learning (RL) has shown promise for decision-making tasks in real-world applications. One practical framework involves training parameterized policy models from an offline dataset and subsequently deploying them in an online environment. However, this approach can be risky since the offline training may not be perfect, leading to poor performance of the RL models that may take dangerous actions. To address this issue, we propose an alternative framework that involves a human supervising the RL models and providing additional feedback in the online deployment phase. We formalize this online deployment problem and develop two approaches. The first approach uses model selection and the upper confidence bound algorithm to adaptively select a model to deploy from a candidate set of trained offline RL models. The second approach involves fine-tuning the model in the online deployment phase when a supervision signal arrives. We demonstrate the effectiveness of these approaches for robot locomotion control and traffic light control tasks through empirical validation.

Every novel technology adds hidden vulnerabilities ready to be exploited by a growing number of cyber-attacks. Automated software testing can be a promising solution to quickly analyze thousands of lines of code by generating and slightly modifying function-specific testing data to encounter a multitude of vulnerabilities and attack vectors. This process draws similarities to the constrained adversarial examples generated by adversarial learning methods, so there could be significant benefits to the integration of these methods in automated testing tools. Therefore, this systematic review is focused on the current state-of-the-art of constrained data generation methods applied for adversarial learning and software testing, aiming to guide researchers and developers to enhance testing tools with adversarial learning methods and improve the resilience and robustness of their digital systems. The found constrained data generation applications for adversarial machine learning were systematized, and the advantages and limitations of approaches specific for software testing were thoroughly analyzed, identifying research gaps and opportunities to improve testing tools with adversarial attack methods.

The ability to learn continually is essential in a complex and changing world. In this paper, we characterize the behavior of canonical value-based deep reinforcement learning (RL) approaches under varying degrees of non-stationarity. In particular, we demonstrate that deep RL agents lose their ability to learn good policies when they cycle through a sequence of Atari 2600 games. This phenomenon is alluded to in prior work under various guises -- e.g., loss of plasticity, implicit under-parameterization, primacy bias, and capacity loss. We investigate this phenomenon closely at scale and analyze how the weights, gradients, and activations change over time in several experiments with varying dimensions (e.g., similarity between games, number of games, number of frames per game), with some experiments spanning 50 days and 2 billion environment interactions. Our analysis shows that the activation footprint of the network becomes sparser, contributing to the diminishing gradients. We investigate a remarkably simple mitigation strategy -- Concatenated ReLUs (CReLUs) activation function -- and demonstrate its effectiveness in facilitating continual learning in a changing environment.