Multi-agent reinforcement learning has demonstrated significant potential in addressing complex cooperative tasks across various real-world applications. However, existing MARL approaches often rely on the restrictive assumption that the number of entities (e.g., agents, obstacles) remains constant between training and inference. This overlooks scenarios where entities are dynamically removed or added during the inference trajectory -- a common occurrence in real-world environments like search and rescue missions and dynamic combat situations. In this paper, we tackle the challenge of intra-trajectory dynamic entity composition under zero-shot out-of-domain (OOD) generalization, where such dynamic changes cannot be anticipated beforehand. Our empirical studies reveal that existing MARL methods suffer significant performance degradation and increased uncertainty in these scenarios. In response, we propose FlickerFusion, a novel OOD generalization method that acts as a universally applicable augmentation technique for MARL backbone methods. FlickerFusion stochastically drops out parts of the observation space, emulating being in-domain when inferenced OOD. The results show that FlickerFusion not only achieves superior inference rewards but also uniquely reduces uncertainty vis-\`a-vis the backbone, compared to existing methods. Benchmarks, implementations, and model weights are organized and open-sourced at flickerfusion305.github.io, accompanied by ample demo video renderings.

Machine learning systems are increasingly being used in critical decision making such as healthcare, finance, and criminal justice. Concerns around their fairness have resulted in several bias mitigation techniques that emphasize the need for high-quality data to ensure fairer decisions. However, the role of earlier stages of machine learning pipelines in mitigating model bias has not been explored well. In this paper, we focus on the task of acquiring additional labeled data points for training the downstream machine learning model to rapidly improve its fairness. Since not all data points in a data pool are equally beneficial to the task of fairness, we generate an ordering in which data points should be acquired. We present DataSift, a data acquisition framework based on the idea of data valuation that relies on partitioning and multi-armed bandits to determine the most valuable data points to acquire. Over several iterations, DataSift selects a partition and randomly samples a batch of data points from the selected partition, evaluates the benefit of acquiring the batch on model fairness, and updates the utility of partitions depending on the benefit. To further improve the effectiveness and efficiency of evaluating batches, we leverage influence functions that estimate the effect of acquiring a batch without retraining the model. We empirically evaluate DataSift on several real-world and synthetic datasets and show that the fairness of a machine learning model can be significantly improved even while acquiring a few data points.

The discipline of automatic control is making increased use of concepts that originate from the domain of machine learning. Herein, reinforcement learning (RL) takes an elevated role, as it is inherently designed for sequential decision making, and can be applied to optimal control problems without the need for a plant system model. To advance education of control engineers and operators in this field, this contribution targets an RL framework that can be applied to educational hardware provided by the Lucas-N\"ulle company. Specifically, the goal of inverted pendulum control is pursued by means of RL, including both, swing-up and stabilization within a single holistic design approach. Herein, the actual learning is enabled by separating corresponding computations from the real-time control computer and outsourcing them to a different hardware. This distributed architecture, however, necessitates communication of the involved components, which is realized via CAN bus. The experimental proof of concept is presented with an applied safeguarding algorithm that prevents the plant from being operated harmfully during the trial-and-error training phase.

Perception components in autonomous systems are often developed and optimized independently of downstream decision-making and control components, relying on established performance metrics like accuracy, precision, and recall. Traditional loss functions, such as cross-entropy loss and negative log-likelihood, focus on reducing misclassification errors but fail to consider their impact on system-level safety, overlooking the varying severities of system-level failures caused by these errors. To address this limitation, we propose a novel training paradigm that augments the perception component with an understanding of system-level safety objectives. Central to our approach is the translation of system-level safety requirements, formally specified using the rulebook formalism, into safety scores. These scores are then incorporated into the reward function of a reinforcement learning framework for fine-tuning perception models with system-level safety objectives. Simulation results demonstrate that models trained with this approach outperform baseline perception models in terms of system-level safety.

We introduce UniGraspTransformer, a universal Transformer-based network for dexterous robotic grasping that simplifies training while enhancing scalability and performance. Unlike prior methods such as UniDexGrasp++, which require complex, multi-step training pipelines, UniGraspTransformer follows a streamlined process: first, dedicated policy networks are trained for individual objects using reinforcement learning to generate successful grasp trajectories; then, these trajectories are distilled into a single, universal network. Our approach enables UniGraspTransformer to scale effectively, incorporating up to 12 self-attention blocks for handling thousands of objects with diverse poses. Additionally, it generalizes well to both idealized and real-world inputs, evaluated in state-based and vision-based settings. Notably, UniGraspTransformer generates a broader range of grasping poses for objects in various shapes and orientations, resulting in more diverse grasp strategies. Experimental results demonstrate significant improvements over state-of-the-art, UniDexGrasp++, across various object categories, achieving success rate gains of 3.5%, 7.7%, and 10.1% on seen objects, unseen objects within seen categories, and completely unseen objects, respectively, in the vision-based setting. Project page: https://dexhand.github.io/UniGraspTransformer.

Closed-loop performance of sequential decision making algorithms, such as model predictive control, depends strongly on the parameters of cost functions, models, and constraints. Bayesian optimization is a common approach to learning these parameters based on closed-loop experiments. However, traditional Bayesian optimization approaches treat the learning problem as a black box, ignoring valuable information and knowledge about the structure of the underlying problem, resulting in slow convergence and high experimental resource use. We propose a time-series-informed optimization framework that incorporates intermediate performance evaluations from early iterations of each experimental episode into the learning procedure. Additionally, probabilistic early stopping criteria are proposed to terminate unpromising experiments, significantly reducing experimental time. Simulation results show that our approach achieves baseline performance with approximately half the resources. Moreover, with the same resource budget, our approach outperforms the baseline in terms of final closed-loop performance, highlighting its efficiency in sequential decision making scenarios.

Inverse Reinforcement Learning (IRL) has demonstrated effectiveness in a variety of imitation tasks. In this paper, we introduce an IRL framework designed to extract rewarding features from expert trajectories affected by delayed disturbances. Instead of relying on direct observations, our approach employs an efficient off-policy adversarial training framework to derive expert features and recover optimal policies from augmented delayed observations. Empirical evaluations in the MuJoCo environment under diverse delay settings validate the effectiveness of our method. Furthermore, we provide a theoretical analysis showing that recovering expert policies from augmented delayed observations outperforms using direct delayed observations.

In this paper, we study the whole-body loco-manipulation problem using reinforcement learning (RL). Specifically, we focus on the problem of how to coordinate the floating base and the robotic arm of a wheeled-quadrupedal manipulator robot to achieve direct six-dimensional (6D) end-effector (EE) pose tracking in task space. Different from conventional whole-body loco-manipulation problems that track both floating-base and end-effector commands, the direct EE pose tracking problem requires inherent balance among redundant degrees of freedom in the whole-body motion. We leverage RL to solve this challenging problem. To address the associated difficulties, we develop a novel reward fusion module (RFM) that systematically integrates reward terms corresponding to different tasks in a nonlinear manner. In such a way, the inherent multi-stage and hierarchical feature of the loco-manipulation problem can be carefully accommodated. By combining the proposed RFM with the a teacher-student RL training paradigm, we present a complete RL scheme to achieve 6D EE pose tracking for the wheeled-quadruped manipulator robot. Extensive simulation and hardware experiments demonstrate the significance of the RFM. In particular, we enable smooth and precise tracking performance, achieving state-of-the-art tracking position error of less than 5 cm, and rotation error of less than 0.1 rad. Please refer to https://clearlab-sustech.github.io/RFM_loco_mani/ for more experimental videos.

The proliferation of Connected Automated Vehicles represents an unprecedented opportunity for improving driving efficiency and alleviating traffic congestion. However, existing research fails to address realistic multi-lane highway scenarios without assuming connectivity, perception, and control capabilities that are typically unavailable in current vehicles. This paper proposes a novel AI system that is the first to improve highway traffic efficiency compared with human-like traffic in realistic, simulated multi-lane scenarios, while relying on existing connectivity, perception, and control capabilities. At the core of our approach is a reinforcement learning based controller that dynamically communicates time-headways to automated vehicles near bottlenecks based on real-time traffic conditions. These desired time-headways are then used by Adaptive Cruise Control (ACC) systems to adjust their following distance. By (i) integrating existing traffic estimation technology and low-bandwidth vehicle-to-infrastructure connectivity, (ii) leveraging safety-certified ACC systems, and (iii) targeting localized bottleneck challenges that can be addressed independently in different locations, we propose a practical, safe, and scalable system that can positively impact numerous road users.

Assistive embodied agents that can be instructed in natural language to perform tasks in open-world environments have the potential to significantly impact labor tasks like manufacturing or in-home care -- benefiting the lives of those who come to depend on them. In this work, we consider how this benefit might be hijacked by local modifications in the appearance of the agent's operating environment. Specifically, we take the popular Vision-and-Language Navigation (VLN) task as a representative setting and develop a whitebox adversarial attack that optimizes a 3D attack object's appearance to induce desired behaviors in pretrained VLN agents that observe it in the environment. We demonstrate that the proposed attack can cause VLN agents to ignore their instructions and execute alternative actions after encountering the attack object -- even for instructions and agent paths not considered when optimizing the attack. For these novel settings, we find our attacks can induce early-termination behaviors or divert an agent along an attacker-defined multi-step trajectory. Under both conditions, environmental attacks significantly reduce agent capabilities to successfully follow user instructions.