Federated Learning (FL) over wireless network enables data-conscious services by leveraging the ubiquitous intelligence at network edge for privacy-preserving model training. As the proliferation of context-aware services, the diversified personal preferences causes disagreeing conditional distributions among user data, which leads to poor inference performance. In this sense, clustered federated learning is proposed to group user devices with similar preference and provide each cluster with a personalized model. This calls for innovative design in edge association that involves user clustering and also resource management optimization. We formulate an accuracy-cost trade-off optimization problem by jointly considering model accuracy, communication resource allocation and energy consumption. To comply with parameter encryption techniques in FL, we propose an iterative solution procedure which employs deep reinforcement learning based approach at cloud server for edge association. The reward function consists of minimized energy consumption at each base station and the averaged model accuracy of all users. Under our proposed solution, multiple edge base station are fully exploited to realize cost efficient personalized federated learning without any prior knowledge on model parameters. Simulation results show that our proposed strategy outperforms existing strategies in achieving accurate learning at low energy cost.

Nowadays, fast delivery services have created the need for high-density warehouses. The puzzle-based storage system is a practical way to enhance the storage density, however, facing difficulties in the retrieval process. In this work, a deep reinforcement learning algorithm, specifically the Double&Dueling Deep Q Network, is developed to solve the multi-item retrieval problem in the system with general settings, where multiple desired items, escorts, and I/O points are placed randomly. Additionally, we propose a general compact integer programming model to evaluate the solution quality. Extensive numerical experiments demonstrate that the reinforcement learning approach can yield high-quality solutions and outperforms three related state-of-the-art heuristic algorithms. Furthermore, a conversion algorithm and a decomposition framework are proposed to handle simultaneous movement and large-scale instances respectively, thus improving the applicability of the PBS system.

Distributional reinforcement learning (RL) aims to learn a value-network that predicts the full distribution of the returns for a given state, often modeled via a quantile-based critic. This approach has been successfully integrated into common RL methods for continuous control, giving rise to algorithms such as Distributional Soft Actor-Critic (DSAC). In this paper, we introduce multi-sample target values (MTV) for distributional RL, as a principled replacement for single-sample target value estimation, as commonly employed in current practice. The improved distributional estimates further lend themselves to UCB-based exploration. These two ideas are combined to yield our distributional RL algorithm, E2DC (Extra Exploration with Distributional Critics). We evaluate our approach on a range of continuous control tasks and demonstrate state-of-the-art model-free performance on difficult tasks such as Humanoid control. We provide further insight into the method via visualization and analysis of the learned distributions and their evolution during training.

Since about 2 years, reinforcement learning has become a hobby for me. I especially enjoy training agents on games. A huge problem for me during these years, was the lack of a reliable reinforcement learning library for python and I had to either program the state of the art algorithms by myself or find a good source on github.

The state of artificial intelligence technology has a rich history that dates back decades and includes two fall-outs before the explosive resurgence of today, which is credited largely to data-driven techniques. While AI technology has and continues to become increasingly mainstream with impact across domains and industries, it's not without several drawbacks, weaknesses, and potential to cause undesired effects. AI techniques are numerous with many approaches and variants, but they can be classified simply based on the degree of knowledge they capture and how much data they require; two broad categories emerge as prominent across AI to date: (1) techniques that are primarily, and often solely, data-driven while leveraging little to no knowledge and (2) techniques that primarily leverage knowledge and depend less on data. Now, a third category is starting to emerge that leverages both data and knowledge, that some refer to as "informed AI." This third category can be a game changer within the national security domain where there is ample scientific and domain-specific knowledge that stands ready to be leveraged, and where purely data-driven AI can lead to serious unwanted consequences. This report shares findings from a thorough exploration of AI approaches that exploit data as well as principled and/or practical knowledge, which we refer to as "knowledge-integrated informed AI." Specifically, we review illuminating examples of knowledge integrated in deep learning and reinforcement learning pipelines, taking note of the performance gains they provide. We also discuss an apparent trade space across variants of knowledge-integrated informed AI, along with observed and prominent issues that suggest worthwhile future research directions. Most importantly, this report suggests how the advantages of knowledge-integrated informed AI stand to benefit the national security domain.

We propose State Matching Offline DIstribution Correction Estimation (SMODICE), a novel and versatile algorithm for offline imitation learning (IL) via state-occupancy matching. We show that the SMODICE objective admits a simple optimization procedure through an application of Fenchel duality and an analytic solution in tabular MDPs. Without requiring access to expert actions, SMODICE can be effectively applied to three offline IL settings: (i) imitation from observations (IfO), (ii) IfO with dynamics or morphologically mismatched expert, and (iii) example-based reinforcement learning, which we show can be formulated as a state-occupancy matching problem. We extensively evaluate SMODICE on both gridworld environments as well as on high-dimensional offline benchmarks. Our results demonstrate that SMODICE is effective for all three problem settings and significantly outperforms prior state-of-art.

Reinforcement learning (RL) is a popular approach for robotic path planning in uncertain environments. However, the control policies trained for an RL agent crucially depend on user-defined, state-based reward functions. Poorly designed rewards can lead to policies that do get maximal rewards but fail to satisfy desired task objectives or are unsafe. There are several examples of the use of formal languages such as temporal logics and automata to specify high-level task specifications for robots (in lieu of Markovian rewards). Recent efforts have focused on inferring state-based rewards from formal specifications; here, the goal is to provide (probabilistic) guarantees that the policy learned using RL (with the inferred rewards) satisfies the high-level formal specification. A key drawback of several of these techniques is that the rewards that they infer are sparse: the agent receives positive rewards only upon completion of the task and no rewards otherwise. This naturally leads to poor convergence properties and high variance during RL. In this work, we propose using formal specifications in the form of symbolic automata: these serve as a generalization of both bounded-time temporal logic-based specifications as well as automata. Furthermore, our use of symbolic automata allows us to define non-sparse potential-based rewards which empirically shape the reward surface, leading to better convergence during RL. We also show that our potential-based rewarding strategy still allows us to obtain the policy that maximizes the satisfaction of the given specification.

The policy gradient theorem (Sutton et al., 2000) prescribes the usage of a cumulative discounted state distribution under the target policy to approximate the gradient. Most algorithms based on this theorem, in practice, break this assumption, introducing a distribution shift that can cause the convergence to poor solutions. In this paper, we propose a new approach of reconstructing the policy gradient from the start state without requiring a particular sampling strategy. The policy gradient calculation in this form can be simplified in terms of a gradient critic, which can be recursively estimated due to a new Bellman equation of gradients. By using temporal-difference updates of the gradient critic from an off-policy data stream, we develop the first estimator that sidesteps the distribution shift issue in a model-free way. We prove that, under certain realizability conditions, our estimator is unbiased regardless of the sampling strategy. We empirically show that our technique achieves a superior bias-variance trade-off and performance in presence of off-policy samples.

Re-configurable robots potentially have more utility and flexibility for many real-world tasks. Designing a learning agent to operate such robots requires adapting to different configurations. While deep reinforcement learning has had immense success in robotic manipulation, domain adaptation is a significant problem that limits its applicability to real-world robotics. We focus on robotic arms with multiple rigid links connected by joints. Recent attempts have performed domain adaptation and Sim2Real transfer to provide robustness to robotic arm dynamics and sensor/camera variations. However, there have been no previous attempts to adapt to robotic arms with a varying number of links. We propose an RL agent with sequence neural networks embedded in the deep neural network to adapt to robotic arms that have a varying number of links. Further, with the additional tool of domain randomization, this agent adapts to different configurations with varying number/length of links and dynamics noise. We perform simulations on a 2D N-link arm to show the ability of our network to transfer and generalize efficiently.

Long-horizon predictions of (top) the Trajectory Transformer compared to those of (bottom) a single-step dynamics model. Modern machine learning success stories often have one thing in common: they use methods that scale gracefully with ever-increasing amounts of data. This is particularly clear from recent advances in sequence modeling, where simply increasing the size of a stable architecture and its training set leads to qualitatively different capabilities.1 Meanwhile, the situation in reinforcement learning has proven more complicated. While it has been possible to apply reinforcement learning algorithms to large–scale problems, generally there has been much more friction in doing so.