In large-scale multi-agent systems, the large number of agents and complex game relationship cause great difficulty for policy learning. Therefore, simplifying the learning process is an important research issue. In many multi-agent systems, the interactions between agents often happen locally, which means that agents neither need to coordinate with all other agents nor need to coordinate with others all the time. Traditional methods attempt to use pre-defined rules to capture the interaction relationship between agents. However, the methods cannot be directly used in a large-scale environment due to the difficulty of transforming the complex interactions between agents into rules. In this paper, we model the relationship between agents by a complete graph and propose a novel game abstraction mechanism based on two-stage attention network (G2ANet), which can indicate whether there is an interaction between two agents and the importance of the interaction. We integrate this detection mechanism into graph neural network-based multi-agent reinforcement learning for conducting game abstraction and propose two novel learning algorithms GA-Comm and GA-AC. We conduct experiments in Traffic Junction and Predator-Prey. The results indicate that the proposed methods can simplify the learning process and meanwhile get better asymptotic performance compared with state-of-the-art algorithms.

Route planning for multiple Unmanned Aerial Vehicles (UAVs) is a series of translation and rotational steps from a given start location to the destination goal location. The goal of the route planning problem is to determine the most optimal route avoiding any collisions with the obstacles present in the environment. Route planning is an NP-hard optimization problem. In this paper, a newly proposed Salp Swarm Algorithm (SSA) is used, and its performance is compared with deterministic and other Nature-Inspired Algorithms (NIAs). The results illustrate that SSA outperforms all the other meta-heuristic algorithms in route planning for multiple UAVs in a 3D environment. The proposed approach improves the average cost and overall time by 1.25% and 6.035% respectively when compared to recently reported data. Route planning is involved in many real-life applications like robot navigation, self-driving car, autonomous UAV for search and rescue operations in dangerous ground-zero situations, civilian surveillance, military combat and even commercial services like package delivery by drones.

Recent years have witnessed significant advances in reinforcement learning (RL), which has registered great success in solving various sequential decision-making problems in machine learning. Most of the successful RL applications, e.g., the games of Go and Poker, robotics, and autonomous driving, involve the participation of more than one single agent, which naturally fall into the realm of multi-agent RL (MARL), a domain with a relatively long history, and has recently re-emerged due to advances in single-agent RL techniques. Though empirically successful, theoretical foundations for MARL are relatively lacking in the literature. In this chapter, we provide a selective overview of MARL, with focus on algorithms backed by theoretical analysis. More specifically, we review the theoretical results of MARL algorithms mainly within two representative frameworks, Markov/stochastic games and extensive-form games, in accordance with the types of tasks they address, i.e., fully cooperative, fully competitive, and a mix of the two. We also introduce several significant but challenging applications of these algorithms. Orthogonal to the existing reviews on MARL, we highlight several new angles and taxonomies of MARL theory, including learning in extensive-form games, decentralized MARL with networked agents, MARL in the mean-field regime, (non-)convergence of policy-based methods for learning in games, etc. Some of the new angles extrapolate from our own research endeavors and interests. Our overall goal with this chapter is, beyond providing an assessment of the current state of the field on the mark, to identify fruitful future research directions on theoretical studies of MARL. We expect this chapter to serve as continuing stimulus for researchers interested in working on this exciting while challenging topic.

Last March, McDonald's Corp. acquired the startup Dynamic Yield for $300 million, in the hope of employing machine learning to personalize customer experience. In the age of artificial intelligence, this was a no-brainer for McDonald's, since Dynamic Yield is widely recognized for its AI-powered technology and recently even landed a spot in a prestigious list of top AI startups. Neural McNetworks are upon us. Trouble is, Dynamic Yield's platform has nothing to do with AI, according to an article posted on Medium last month by the company's former head of content, Mike Mallazzo. It was a heartfelt takedown of phony AI, which was itself taken down by the author but remains engraved in the collective memory of the internet.

This paper proposes a class of distributed event-triggered algorithms that solve the average consensus problem in multi-agent systems. By designing events such that a specifically chosen Lyapunov function is monotonically decreasing, event-triggered algorithms succeed in reducing communications among agents while still ensuring that the entire system converges to the desired state. However, depending on the chosen Lyapunov function the transient behaviors can be very different. Moreover, performance requirements also vary from application to application. Consequently, we are instead interested in considering a class of Lyapunov functions such that each Lyapunov function produces a different event-triggered coordination algorithm to solve the multi-agent average consensus problem. The proposed class of algorithms all guarantee exponential convergence of the resulting system and exclusion of Zeno behaviors. This allows us to easily implement different algorithms that all guarantee correctness to meet varying performance needs. We show that our findings can be applied to the practical clock synchronization problem in wireless sensor networks (WSNs) and further corroborate their effectiveness with simulation results.

We are the Department of Data Science and Knowledge Engineering (DKE) at Maastricht University, the Netherlands: an international community of 50 researchers at various stages of their career, embedded in the Faculty of Science and Engineering (FSE). Our department has nearly 30 years' experience with research and teaching in the fields of Artificial Intelligence, Computer Science and Mathematics, and we do so in a highly collaborative and cross-disciplinary manner. To strengthen our team, we are looking for a full professor who will work on AI systems that interact intelligently – and in a human-centered and responsible way - with other AI systems, humans, or both. Our department is growing rapidly. This position is one of multiple job openings: you are more than welcome to browse through our other vacancies.

Often, closed -- form analytic formulations are not available and forecasting or decision making tasks have to rely on noisy, irregularly sampled observations. This class of systems offers a crystal clear example of inductive relational biases. Introducing inductive biases in statistics or machine learning is a well known approach to improving sample efficiency and generalization performance. From the choice of objective function, to the design of ad -- hoc deep learning architectures suited to the specific problem at hand, biases are common and effective. Relational inductive biases [1] represent a special class of biases, concerned with relationship between entities.

Multi-agent coordination is prevalent in many real-world applications. However, such coordination is challenging due to its combinatorial nature. An important observation in this regard is that agents in the real world often only directly affect a limited set of neighboring agents. Leveraging such loose couplings among agents is key to making coordination in multi-agent systems feasible. In this work, we focus on learning to coordinate. Specifically, we consider the multi-agent multi-armed bandit framework, in which fully cooperative loosely-coupled agents must learn to coordinate their decisions to optimize a common objective. As opposed to in the planning setting, for learning methods it is challenging to establish theoretical guarantees. We propose multi-agent Thompson sampling (MATS), a new Bayesian exploration-exploitation algorithm that leverages loose couplings. We provide a regret bound that is sublinear in time and low-order polynomial in the highest number of actions of a single agent for sparse coordination graphs. Finally, we empirically show that MATS outperforms the state-of-the-art algorithm, MAUCE, on two synthetic benchmarks, a realistic wind farm control task, and a novel benchmark with Poisson distributions.

We present CoverNet, a new method for multimodal, probabilistic trajectory prediction in urban driving scenarios. Previous work has employed a variety of methods, including multimodal regression, occupancy maps, and 1-step stochastic policies. We instead frame the trajectory prediction problem as classification over a diverse set of trajectories. The size of this set remains manageable, due to the fact that there are a limited number of distinct actions that can be taken over a reasonable prediction horizon. We structure the trajectory set to a) ensure a desired level of coverage of the state space, and b) eliminate physically impossible trajectories. By dynamically generating trajectory sets based on the agent's current state, we can further improve the efficiency of our method. We demonstrate our approach on public, real-world self-driving datasets, and show that it outperforms state-of-the-art methods.

The paper provides an understanding of social capital in organizations that are open membership multi-agent systems with an emphasis in our formulation on the dynamic network of social interaction that, in part, elucidate evolving structures and impromptu topologies of networks. This paper, therefore, models an open source project as an organizational network. It provides definitions of social capital for this organizational network and formulation of the mechanism to optimize the social capital for achieving its goal that is optimized productivity. A case study of an open source Apache-Hadoop project is considered and empirically evaluated. An analysis of how social capital can be created within this type of organizations and driven to a measurement for its value is provided. Finally, a verification on whether the social capital of the organizational network is proportional towards optimizing their productivity is considered.