This paper addresses the considerations that comes along with adopting decentralized communication for multi-agent localization applications in discrete state spaces. In this framework, we extend the original formulation of the Bayes filter, a foundational probabilistic tool for discrete state estimation, by appending a step of greedy belief sharing as a method to propagate information and improve local estimates' posteriors. We apply our work in a model-based multi-agent grid-world setting, where each agent maintains a belief distribution for every agents' state. Our results affirm the utility of our proposed extensions for decentralized collaborative tasks.

Wide machine learning tasks can be formulated as non-convex multi-player games, where Nash equilibrium (NE) is an acceptable solution to all players, since no one can benefit from changing its strategy unilaterally. Attributed to the non-convexity, obtaining the existence condition of global NE is challenging, let alone designing theoretically guaranteed realization algorithms. This paper takes conjugate transformation to the formulation of non-convex multi-player games, and casts the complementary problem into a variational inequality (VI) problem with a continuous pseudo-gradient mapping. We then prove the existence condition of global NE: the solution to the VI problem satisfies a duality relation. Based on this VI formulation, we design a conjugate-based ordinary differential equation (ODE) to approach global NE, which is proved to have an exponential convergence rate. To make the dynamics more implementable, we further derive a discretized algorithm. We apply our algorithm to two typical scenarios: multi-player generalized monotone game and multi-player potential game. In the two settings, we prove that the step-size setting is required to be $\mathcal{O}(1/k)$ and $\mathcal{O}(1/\sqrt k)$ to yield the convergence rates of $\mathcal{O}(1/ k)$ and $\mathcal{O}(1/\sqrt k)$, respectively. Extensive experiments in robust neural network training and sensor localization are in full agreement with our theory.

Multi-Agent RL or MARL is one of the complex problems in Autonomous Driving literature that hampers the release of fully-autonomous vehicles today. Several simulators have been in iteration after their inception to mitigate the problem of complex scenarios with multiple agents in Autonomous Driving. One such simulator--SMARTS, discusses the importance of cooperative multi-agent learning. For this problem, we discuss two approaches--MAPPO and MADDPG, which are based on-policy and off-policy RL approaches. We compare our results with the state-of-the-art results for this challenge and discuss the potential areas of improvement while discussing the explainability of these approaches in conjunction with waypoints in the SMARTS environment.

Decentralized bilevel optimization has received increasing attention recently due to its foundational role in many emerging multi-agent learning paradigms (e.g., multi-agent meta-learning and multi-agent reinforcement learning) over peer-to-peer edge networks. However, to work with the limited computation and communication capabilities of edge networks, a major challenge in developing decentralized bilevel optimization techniques is to lower sample and communication complexities. This motivates us to develop a new decentralized bilevel optimization called DIAMOND (decentralized single-timescale stochastic approximation with momentum and gradient-tracking). The contributions of this paper are as follows: i) our DIAMOND algorithm adopts a single-loop structure rather than following the natural double-loop structure of bilevel optimization, which offers low computation and implementation complexity; ii) compared to existing approaches, the DIAMOND algorithm does not require any full gradient evaluations, which further reduces both sample and computational complexities; iii) through a careful integration of momentum information and gradient tracking techniques, we show that the DIAMOND algorithm enjoys $\mathcal{O}(\epsilon^{-3/2})$ in sample and communication complexities for achieving an $\epsilon$-stationary solution, both of which are independent of the dataset sizes and significantly outperform existing works. Extensive experiments also verify our theoretical findings.

Several approaches have been proposed and studied, such as Intrinsic Motivation, co-evolution of agents and tasks, and multi-agent competition. In particular, the interplay between multiple agents has proven to be capable of generating human-relevant emergent behaviour that would be difficult or impossible to learn in single-agent settings. In this work, an extensible competitive environment for multi-agent interplay was developed, which features realistic physics and human-relevant semantics. Moreover, several experiments on different variants of this environment were performed, resulting in some simple emergent strategies and concrete directions for future improvement. The content presented here is part of the author's thesis "Multi-Agent Interplay in a Competitive Survival Environment" for the Master's Degree in Artificial Intelligence and Robotics at Sapienza University of Rome, 2022.

The field of emergent communication aims to understand the characteristics of communication as it emerges from artificial agents solving tasks that require information exchange. Communication with discrete messages is considered a desired characteristic, for both scientific and applied reasons. However, training a multi-agent system with discrete communication is not straightforward, requiring either reinforcement learning algorithms or relaxing the discreteness requirement via a continuous approximation such as the Gumbel-softmax. Both these solutions result in poor performance compared to fully continuous communication. In this work, we propose an alternative approach to achieve discrete communication -- quantization of communicated messages. Using message quantization allows us to train the model end-to-end, achieving superior performance in multiple setups. Moreover, quantization is a natural framework that runs the gamut from continuous to discrete communication. Thus, it sets the ground for a broader view of multi-agent communication in the deep learning era.

By incorporating ergonomics principles into the task allocation processes, human-robot collaboration (HRC) frameworks can favour the prevention of work-related musculoskeletal disorders (WMSDs). In this context, existing offline methodologies do not account for the variability of human actions and states; therefore, planning and dynamically assigning roles in human-robot teams remains an unaddressed challenge.This study aims to create an ergonomic role allocation framework that optimises the HRC, taking into account task features and human state measurements. The presented framework consists of two main modules: the first provides the HRC task model, exploiting AND/OR Graphs (AOG)s, which we adapted to solve the allocation problem; the second module describes the ergonomic risk assessment during task execution through a risk indicator and updates the AOG-related variables to influence future task allocation. The proposed framework can be combined with any time-varying ergonomic risk indicator that evaluates human cognitive and physical burden. In this work, we tested our framework in an assembly scenario, introducing a risk index named Kinematic Wear.The overall framework has been tested with a multi-subject experiment. The task allocation results and subjective evaluations, measured with questionnaires, show that high-risk actions are correctly recognised and not assigned to humans, reducing fatigue and frustration in collaborative tasks.

Practical deployments of coordinated fleets of mobile robots in different environments have revealed the benefits of maintaining small distances between robots, especially as they move at higher speeds. However, this is counter-intuitive in that as speed increases, reducing the amount of space between robots also reduces the time available to the robots to respond to sudden motion variations in surrounding robots. However, in certain examples, the benefits in performance due to traveling at closer distances can outweigh the potential instability issues, for instance, autonomous trucks on highways that optimize energy by vehicle ``drafting'' or smaller robots in cluttered environments that need to maintain close, line of sight communication, etc. To achieve this kind of closely coordinated fleet behavior, this work introduces a model predictive optimal control framework that directly takes non-linear dynamics of the vehicles in the fleet into account while planning motions for each robot. The robots are able to follow each other closely at high speeds by proactively making predictions and reactively biasing their responses based on state information from the adjacent robots. This control framework is naturally decentralized and, as such, is able to apply to an arbitrary number of robots without any additional computational burden. We show that our approach is able to achieve lower inter-robot distances at higher speeds compared to existing controllers. We demonstrate the success of our approach through simulated and hardware results on mobile ground robots.

Enabling autonomous agents to act cooperatively is an important step to integrate artificial intelligence in our daily lives. While some methods seek to stimulate cooperation by letting agents give rewards to others, in this paper we propose a method inspired by the stock market, where agents have the opportunity to participate in other agents' returns by acquiring reward shares. Intuitively, an agent may learn to act according to the common interest when being directly affected by the other agents' rewards. The empirical results of the tested general-sum Markov games show that this mechanism promotes cooperative policies among independently trained agents in social dilemma situations. Moreover, as demonstrated in a temporally and spatially extended domain, participation can lead to the development of roles and the division of subtasks between the agents.

Foundation models have shown impressive adaptation and scalability in supervised and self-supervised learning problems, but so far these successes have not fully translated to reinforcement learning (RL). In this work, we demonstrate that training an RL agent at scale leads to a general in-context learning algorithm that can adapt to open-ended novel embodied 3D problems as quickly as humans. In a vast space of held-out environment dynamics, our adaptive agent (AdA) displays on-the-fly hypothesis-driven exploration, efficient exploitation of acquired knowledge, and can successfully be prompted with first-person demonstrations. Adaptation emerges from three ingredients: (1) meta-reinforcement learning across a vast, smooth and diverse task distribution, (2) a policy parameterised as a large-scale attention-based memory architecture, and (3) an effective automated curriculum that prioritises tasks at the frontier of an agent's capabilities. We demonstrate characteristic scaling laws with respect to network size, memory length, and richness of the training task distribution. We believe our results lay the foundation for increasingly general and adaptive RL agents that perform well across ever-larger open-ended domains.