Algorithmic agents are used in a variety of competitive decision settings, notably in making pricing decisions in contexts that range from online retail to residential home rentals. Business managers, algorithm designers, legal scholars, and regulators alike are all starting to consider the ramifications of "algorithmic collusion." We study the emergent behavior of multi-armed bandit machine learning algorithms used in situations where agents are competing, but they have no information about the strategic interaction they are engaged in. Using a general-form repeated Prisoner's Dilemma game, agents engage in online learning with no prior model of game structure and no knowledge of competitors' states or actions (e.g., no observation of competing prices). We show that these context-free bandits, with no knowledge of opponents' choices or outcomes, still will consistently learn collusive behavior - what we call "naive collusion." We primarily study this system through an analytical model and examine perturbations to the model through simulations. Our findings have several notable implications for regulators. First, calls to limit algorithms from conditioning on competitors' prices are insufficient to prevent algorithmic collusion. This is a direct result of collusion arising even in the naive setting. Second, symmetry in algorithms can increase collusion potential. This highlights a new, simple mechanism for "hub-and-spoke" algorithmic collusion. A central distributor need not imbue its algorithm with supra-competitive tendencies for apparent collusion to arise; it can simply arise by using certain (common) machine learning algorithms. Finally, we highlight that collusive outcomes depend starkly on the specific algorithm being used, and we highlight market and algorithmic conditions under which it will be unknown a priori whether collusion occurs.

Effective human-AI collaboration requires agents to adopt their roles and levels of support based on human needs, task requirements, and complexity. Traditional human-AI teaming often relies on a pre-determined robot communication scheme, restricting teamwork adaptability in complex tasks. Leveraging the strong communication capabilities of Large Language Models (LLMs), we propose a Human-Robot Teaming Framework with Multi-Modal Language feedback (HRT-ML), a framework designed to enhance human-robot interaction by adjusting the frequency and content of language-based feedback. The HRT-ML framework includes two core modules: a Coordinator for high-level, low-frequency strategic guidance and a Manager for task-specific, high-frequency instructions, enabling passive and active interactions with human teammates. To assess the impact of language feedback in collaborative scenarios, we conducted experiments in an enhanced Overcooked-AI game environment with varying levels of task complexity (easy, medium, hard) and feedback frequency (inactive, passive, active, superactive). Our results show that as task complexity increases relative to human capabilities, human teammates exhibited stronger preferences toward robotic agents that can offer frequent, proactive support. However, when task complexities exceed the LLM's capacity, noisy and inaccurate feedback from superactive agents can instead hinder team performance, as it requires human teammates to increase their effort to interpret and respond to the large amount of communications, with limited performance return. Our results offer a general principle for robotic agents to dynamically adjust their levels and frequencies of communication to work seamlessly with humans and achieve improved teaming performance.

Simultaneous localization and mapping (SLAM) systems with novel view synthesis capabilities are widely used in computer vision, with applications in augmented reality, robotics, and autonomous driving. However, existing approaches are limited to single-agent operation. Recent work has addressed this problem using a distributed neural scene representation. Unfortunately, existing methods are slow, cannot accurately render real-world data, are restricted to two agents, and have limited tracking accuracy. In contrast, we propose a rigidly deformable 3D Gaussian-based scene representation that dramatically speeds up the system. However, improving tracking accuracy and reconstructing a globally consistent map from multiple agents remains challenging due to trajectory drift and discrepancies across agents' observations. Therefore, we propose new tracking and map-merging mechanisms and integrate loop closure in the Gaussian-based SLAM pipeline. We evaluate MAGiC-SLAM on synthetic and real-world datasets and find it more accurate and faster than the state of the art.

We introduce the concept of leakage-robust Bayesian persuasion. Situated between public persuasion [KG11, CCG23, Xu20] and private persuasion [AB19], leakage-robust persuasion considers a setting where one or more signals privately sent by a sender to the receivers may be leaked. We study the design of leakage-robust persuasion schemes and quantify the price of robustness using two formalisms: - The first notion, $k$-worst-case persuasiveness, requires a scheme to remain persuasive as long as each receiver observes at most $k$ leaked signals. We quantify the Price of Worst-case Robustness (PoWR$_k$) -- i.e., the gap in sender's utility as compared to the optimal private scheme -- as $\Theta(\min\{2^k,n\})$ for supermodular sender utilities and $\Theta(k)$ for submodular or XOS utilities, where $n$ is the number of receivers. This result also establishes that in some instances, $\Theta(\log k)$ leakages are sufficient for the utility of the optimal leakage-robust persuasion to degenerate to that of public persuasion. - The second notion, expected downstream utility robustness, relaxes the persuasiveness and considers the impact on sender's utility when receivers best respond to their observations. By quantifying the Price of Downstream Robustness (PoDR) as the gap between the sender's expected utility over random leakage patterns as compared to private persuasion, we show that over several natural and structured distributions of leakage patterns, PoDR improves PoWR to $\Theta(k)$ or even $\Theta(1)$, where $k$ is the maximum number of leaked signals observable to each receiver across leakage patterns in the distribution. En route to these results, we show that subsampling and masking are general-purpose algorithmic paradigms for transforming private persuasion signaling schemes to leakage-robust ones, with minmax optimal loss in the sender's utility.

Motivated by practical federated learning settings where clients may not be always available, In this work we focus on distributed online optimization we investigate a variant of distributed (DOO), an online learning variant of distributed online optimization where agents are active convex optimization in which each agent is facing an with a known probability p at each time adversarial sequence of convex loss functions (Hosseini step, and communication between neighboring et al., 2013). The goal of an agent is to minimize its agents can only take place if they are regret with respect to a sequence of global loss functions, both active. We introduce a distributed variant each obtained by summing the corresponding of the FTRL algorithm and analyze its local losses for each agent. In both batch and online network regret, defined through the average distributed optimization settings, the presence of of the instantaneous regret of the active the communication network, which limits the exchange agents. Our analysis shows that, for any of information to adjacent nodes, implies that agents connected communication graph G over N must use some information-propagation technique to agents, the expected network regret of our collect information about the global loss function.

Ensuring the safety of autonomous vehicles (AVs) requires identifying rare but critical failure cases that on-road testing alone cannot discover. High-fidelity simulations provide a scalable alternative, but automatically generating realistic and diverse traffic scenarios that can effectively stress test AV motion planners remains a key challenge. This paper introduces CRASH - Challenging Reinforcement-learning based Adversarial scenarios for Safety Hardening - an adversarial deep reinforcement learning framework to address this issue. First CRASH can control adversarial Non Player Character (NPC) agents in an AV simulator to automatically induce collisions with the Ego vehicle, falsifying its motion planner. We also propose a novel approach, that we term safety hardening, which iteratively refines the motion planner by simulating improvement scenarios against adversarial agents, leveraging the failure cases to strengthen the AV stack. CRASH is evaluated on a simplified two-lane highway scenario, demonstrating its ability to falsify both rule-based and learning-based planners with collision rates exceeding 90%. Additionally, safety hardening reduces the Ego vehicle's collision rate by 26%. While preliminary, these results highlight RL-based safety hardening as a promising approach for scenario-driven simulation testing for autonomous vehicles.

In this article, we address the problem of designing a scalable control architecture for a safe coordinated operation of a multi-agent system with aerial (UAVs) and ground robots (UGVs) in a confined task space. The proposed method uses Control Barrier Functions (CBFs) to impose constraints associated with (i) collision avoidance between agents, (ii) landing of UAVs on mobile UGVs, and (iii) task space restriction. Further, to account for the rapid increase in the number of constraints for a single agent with the increasing number of agents, the proposed architecture uses a centralized-decentralized Edge cluster, where a centralized node (Watcher) activates the relevant constraints, reducing the need for high onboard processing and network complexity. The distributed nodes run the controller locally to overcome latency and network issues. The proposed Edge architecture is experimentally validated using multiple aerial and ground robots in a confined environment performing a coordinated operation.

Imitation learning in robotics faces significant challenges in generalization due to the complexity of robotic environments and the high cost of data collection. We introduce RoCoDA, a novel method that unifies the concepts of invariance, equivariance, and causality within a single framework to enhance data augmentation for imitation learning. RoCoDA leverages causal invariance by modifying task-irrelevant subsets of the environment state without affecting the policy's output. Simultaneously, we exploit SE(3) equivariance by applying rigid body transformations to object poses and adjusting corresponding actions to generate synthetic demonstrations. We validate RoCoDA through extensive experiments on five robotic manipulation tasks, demonstrating improvements in policy performance, generalization, and sample efficiency compared to state-of-the-art data augmentation methods. Our policies exhibit robust generalization to unseen object poses, textures, and the presence of distractors. Furthermore, we observe emergent behavior such as re-grasping, indicating policies trained with RoCoDA possess a deeper understanding of task dynamics. By leveraging invariance, equivariance, and causality, RoCoDA provides a principled approach to data augmentation in imitation learning, bridging the gap between geometric symmetries and causal reasoning.

An agent trained within a closed system can master any desired capability, as long as the following three conditions hold: (a) it receives sufficiently informative and aligned feedback, (b) its coverage of experience/data is broad enough, and (c) it has sufficient capacity and resource. In this position paper, we justify these conditions, and consider what limitations arise from (a) and (b) in closed systems, when assuming that (c) is not a bottleneck. Considering the special case of agents with matching input and output spaces (namely, language), we argue that such pure recursive self-improvement, dubbed "Socratic learning", can boost performance vastly beyond what is present in its initial data or knowledge, and is only limited by time, as well as gradual misalignment concerns. Furthermore, we propose a constructive framework to implement it, based on the notion of language games.

In this paper, we present a novel trajectory prediction model for autonomous driving, combining a Characterized Diffusion Module and a Spatial-Temporal Interaction Network to address the challenges posed by dynamic and heterogeneous traffic environments. Our model enhances the accuracy and reliability of trajectory predictions by incorporating uncertainty estimation and complex agent interactions. Through extensive experimentation on public datasets such as NGSIM, HighD, and MoCAD, our model significantly outperforms existing state-of-the-art methods. We demonstrate its ability to capture the underlying spatial-temporal dynamics of traffic scenarios and improve prediction precision, especially in complex environments. The proposed model showcases strong potential for application in real-world autonomous driving systems.