Particle robots are novel biologically-inspired robotic systems where locomotion can be achieved collectively and robustly, but not independently. While its control is currently limited to a hand-crafted policy for basic locomotion tasks, such a multi-robot system could be potentially controlled via Deep Reinforcement Learning (DRL) for different tasks more efficiently. However, the particle robot system presents a new set of challenges for DRL differing from existing swarm robotics systems: the low degrees of freedom of each robot and the increased necessity of coordination between robots. We present a 2D particle robot simulator using the OpenAI Gym interface and Pymunk as the physics engine, and introduce new tasks and challenges to research the underexplored applications of DRL in the particle robot system. Moreover, we use Stable-baselines3 to provide a set of benchmarks for the tasks. Current baseline DRL algorithms show signs of achieving the tasks but are yet unable to reach the performance of the hand-crafted policy. Further development of DRL algorithms is necessary in order to accomplish the proposed tasks.

In performative Reinforcement Learning (RL), an agent faces a policy-dependent environment: the reward and transition functions depend on the agent's policy. Prior work on performative RL has studied the convergence of repeated retraining approaches to a performatively stable policy. In the finite sample regime, these approaches repeatedly solve for a saddle point of a convex-concave objective, which estimates the Lagrangian of a regularized version of the reinforcement learning problem. In this paper, we aim to extend such repeated retraining approaches, enabling them to operate under corrupted data. More specifically, we consider Huber's $ε$-contamination model, where an $ε$ fraction of data points is corrupted by arbitrary adversarial noise. We propose a repeated retraining approach based on convex-concave optimization under corrupted gradients and a novel problem-specific robust mean estimator for the gradients. We prove that our approach exhibits last-iterate convergence to an approximately stable policy, with the approximation error linear in $\sqrtε$. We experimentally demonstrate the importance of accounting for corruption in performative RL.

In addition to a policy update, AC methods employ a parallel critic update to bootstrap the Q-value for policy gradient estimation, which often enjoys reduced variance and fast convergence in training. Despite the empirical success, theoretical analysis of AC in the most practical form remains challenging. Existing works mostly focus on either the double-loop or the two-timescale variants. In double-loop AC, the actor is updated in the outer loop only after the critic takes sufficiently many steps to have an accurate estimation of the Q-value in the inner loop [ Y anget al., 2019; Kumar et al., 2019; Wang et al., 2019 ] . Hence, the convergence of the critic is decoupled from that of the actor. The analysis is separated into a policy evaluation sub-problem in the inner loop and a perturbed gradient descent in the outer loop. In two-timescale AC, the actor and the critic are updated simultaneously in each iteration using stepsizes of different timescales. The actor stepsize (denoted by α t in the sequel) is typically smaller than that of the critic (denoted by β t in the sequel), with their ratio going to zero as the iteration number goes to infinity (i.e., lim t α t/β t = 0). The two-timescale allows the critic to approximate the correct Q-value asymptotically.

Interactive imitation learning makes an agent's control policy robust by stepwise supervisions from an expert. The recent algorithms mostly employ expert-agent switching systems to reduce the expert's burden by limitedly selecting the supervision timing. However, the precise selection is difficult and such a switching causes abrupt changes in actions, damaging the dynamic stability. This paper therefore proposes a novel method, so-called CubeDAgger, which improves robustness while reducing dynamic stability violations by making three improvements to a baseline method, EnsembleDAgger. The first improvement adds a regularization to explicitly activate the threshold for deciding the supervision timing. The second transforms the expert-agent switching system to an optimal consensus system of multiple action candidates. Third, autoregressive colored noise to the actions is introduced to make the stochastic exploration consistent over time. These improvements are verified by simulations, showing that the learned policies are sufficiently robust while maintaining dynamic stability during interaction.

DBSCAN, a well-known density-based clustering algorithm, has gained widespread popularity and usage due to its effectiveness in identifying clusters of arbitrary shapes and handling noisy data. However, it encounters challenges in producing satisfactory cluster results when confronted with datasets of varying density scales, a common scenario in real-world applications. In this paper, we propose a novel Adaptive and Robust DBSCAN with Multi-agent Reinforcement Learning cluster framework, namely AR-DBSCAN. First, we model the initial dataset as a two-level encoding tree and categorize the data vertices into distinct density partitions according to the information uncertainty determined in the encoding tree. Each partition is then assigned to an agent to find the best clustering parameters without manual assistance. The allocation is density-adaptive, enabling AR-DBSCAN to effectively handle diverse density distributions within the dataset by utilizing distinct agents for different partitions. Second, a multi-agent deep reinforcement learning guided automatic parameter searching process is designed. The process of adjusting the parameter search direction by perceiving the clustering environment is modeled as a Markov decision process. Using a weakly-supervised reward training policy network, each agent adaptively learns the optimal clustering parameters by interacting with the clusters. Third, a recursive search mechanism adaptable to the data's scale is presented, enabling efficient and controlled exploration of large parameter spaces. Extensive experiments are conducted on nine artificial datasets and a real-world dataset. The results of offline and online tasks show that AR-DBSCAN not only improves clustering accuracy by up to 144.1% and 175.3% in the NMI and ARI metrics, respectively, but also is capable of robustly finding dominant parameters.

Machine learning has become a powerful tool for enhancing data assimilation. While supervised learning remains the standard method, reinforcement learning (RL) offers unique advantages through its sequential decision-making framework, which naturally fits the iterative nature of data assimilation by dynamically balancing model forecasts with observations. We develop RL-DAUNCE, a new RL-based method that enhances data assimilation with physical constraints through three key aspects. First, RL-DAUNCE inherits the computational efficiency of machine learning while it uniquely structures its agents to mirror ensemble members in conventional data assimilation methods. Second, RL-DAUNCE emphasizes uncertainty quantification by advancing multiple ensemble members, moving beyond simple mean-state optimization. Third, RL-DAUNCE's ensemble-as-agents design facilitates the enforcement of physical constraints during the assimilation process, which is crucial to improving the state estimation and subsequent forecasting. A primal-dual optimization strategy is developed to enforce constraints, which dynamically penalizes the reward function to ensure constraint satisfaction throughout the learning process. Also, state variable bounds are respected by constraining the RL action space. Together, these features ensure physical consistency without sacrificing efficiency. RL-DAUNCE is applied to the Madden-Julian Oscillation, an intermittent atmospheric phenomenon characterized by strongly non-Gaussian features and multiple physical constraints. RL-DAUNCE outperforms the standard ensemble Kalman filter (EnKF), which fails catastrophically due to the violation of physical constraints. Notably, RL-DAUNCE matches the performance of constrained EnKF, particularly in recovering intermittent signals, capturing extreme events, and quantifying uncertainties, while requiring substantially less computational effort.

Simplicity is a critical inductive bias for designing data-driven controllers, especially when robustness is important. Despite the impressive results of deep reinforcement learning in complex control tasks, it is prone to capturing intricate and spurious correlations between observations and actions, leading to failure under slight perturbations to the environment. To tackle this problem, in this work we introduce a novel inductive bias towards simple policies in reinforcement learning. The simplicity inductive bias is introduced by minimizing the entropy of entire action trajectories, corresponding to the number of bits required to describe information in action trajectories after the agent observes state trajectories. Our reinforcement learning agent, Trajectory Entropy Reinforcement Learning, is optimized to minimize the trajectory entropy while maximizing rewards. We show that the trajectory entropy can be effectively estimated by learning a variational parameterized action prediction model, and use the prediction model to construct an information-regularized reward function. Furthermore, we construct a practical algorithm that enables the joint optimization of models, including the policy and the prediction model. Experimental evaluations on several high-dimensional locomotion tasks show that our learned policies produce more cyclical and consistent action trajectories, and achieve superior performance, and robustness to noise and dynamic changes than the state-of-the-art.

Coverage Path Planning (CPP) is vital in precision agriculture to improve efficiency and resource utilization. In irregular and dispersed plantations, traditional grid-based CPP often causes redundant coverage over non-vegetated areas, leading to waste and pollution. To overcome these limitations, we propose CPP-DIP, a multi-objective CPP framework designed for Micro Air Vehicles (MAVs). The framework transforms the CPP task into a Traveling Salesman Problem (TSP) and optimizes flight paths by minimizing travel distance, turning angles, and intersection counts. Unlike conventional approaches, our method does not rely on GPS-based environmental modeling. Instead, it uses aerial imagery and a Histogram of Oriented Gradients (HOG)-based approach to detect trees and extract image coordinates. A density-aware waypoint strategy is applied: Kernel Density Estimation (KDE) is used to reduce redundant waypoints in dense regions, while a greedy algorithm ensures complete coverage in sparse areas. To verify the generality of the framework, we solve the resulting TSP using three different methods: Greedy Heuristic Insertion (GHI), Ant Colony Optimization (ACO), and Monte Carlo Reinforcement Learning (MCRL). Then an object-based optimization is applied to further refine the resulting path. Additionally, CPP-DIP integrates ForaNav, our insect-inspired navigation method, for accurate tree localization and tracking. The experimental results show that MCRL offers a balanced solution, reducing the travel distance by 16.9 % compared to ACO while maintaining a similar performance to GHI. It also improves path smoothness by reducing turning angles by 28.3 % and 59.9 % relative to ACO and GHI, respectively, and effectively eliminates intersections. These results confirm the robustness and effectiveness of CPP-DIP in different TSP solvers.

The layout of analog ICs requires making complex trade-offs, while addressing device physics and variability of the circuits. This makes full automation with learning-based solutions hard to achieve. However, reinforcement learning (RL) has recently reached significant results, particularly in solving the floorplanning problem. This paper presents a hybrid method that combines RL with a beam (BS) strategy. The BS algorithm enhances the agent's inference process, allowing for the generation of flexible floorplans by accomodating various objective weightings, and addressing congestion without without the need for policy retraining or fine-tuning. Moreover, the RL agent's generalization ability stays intact, along with its efficient handling of circuit features and constraints. Experimental results show approx. 5-85% improvement in area, dead space and half-perimeter wire length compared to a standard RL application, along with higher rewards for the agent. Moreover, performance and efficiency align closely with those of existing state-of-the-art techniques.

Terahertz (THz) unmanned aerial vehicle (UAV) networks with flexible topologies and ultra-high data rates are expected to empower numerous applications in security surveillance, disaster response, and environmental monitoring, among others. However, the dynamic topologies hinder the efficient long-term joint power and antenna array resource allocation for THz links among UAVs. Furthermore, the continuous nature of power and the discrete nature of antennas cause this joint resource allocation problem to be a mixed-integer nonlinear programming (MINLP) problem with non-convexity and NP-hardness. Inspired by recent rapid advancements in deep reinforcement learning (DRL), a graph neural network (GNN) aided DRL algorithm for resource allocation in the dynamic THz UAV network with an emphasis on self-node features (GLOVE) is proposed in this paper, with the aim of resource efficiency (RE) maximization. When training the allocation policy for each UAV, GLOVE learns the relationship between this UAV and its neighboring UAVs via GNN, while also emphasizing the important self-node features of this UAV. In addition, a multi-task structure is leveraged by GLOVE to cooperatively train resource allocation decisions for the power and sub-arrays of all UAVs. Experimental results illustrate that GLOVE outperforms benchmark schemes in terms of the highest RE and the lowest latency. Moreover, unlike the benchmark methods with severe packet loss, GLOVE maintains zero packet loss during the entire training process, demonstrating its better robustness under the highly dynamic THz UAV network.