Multi-agent interacting systems are prevalent in the world, from purely physical systems to complicated social dynamic systems. The interactions between entities / components can give rise to very complex behavior patterns at the level of both individuals and the multi-agent system as a whole. Since usually only the trajectories of individual entities are observed without any knowledge of the underlying interaction patterns, and there are usually multiple possible modalities for each agent with uncertainty, it is challenging to model their dynamics and forecast their future behaviors. We introduce a generic trajectory forecasting framework (named EvolveGraph) with explicit relational structure recognition and prediction via latent interaction graphs among multiple heterogeneous, interactive agents. Considering the uncertainty of future behaviors, the model is designed to provide multi-modal prediction hypotheses.

Most recently developed approaches to cooperative multi-agent reinforcement learning in the \emph{centralized training with decentralized execution} setting involve estimating a centralized, joint value function. In this paper, we demonstrate that, despite its various theoretical shortcomings, Independent PPO (IPPO), a form of independent learning in which each agent simply estimates its local value function, can perform just as well as or better than state-of-the-art joint learning approaches on popular multi-agent benchmark suite SMAC with little hyperparameter tuning. We also compare IPPO to several variants; the results suggest that IPPO's strong performance may be due to its robustness to some forms of environment non-stationarity.

In the pursuit of artificial general intelligence, our most significant measurement of progress is an agent's ability to achieve goals in a wide range of environments. Existing platforms for constructing such environments are typically constrained by the technologies they are founded on, and are therefore only able to provide a subset of scenarios necessary to evaluate progress. To overcome these shortcomings, we present our use of Unity, a widely recognized and comprehensive game engine, to create more diverse, complex, virtual simulations. We describe the concepts and components developed to simplify the authoring of these environments, intended for use predominantly in the field of reinforcement learning. We also introduce a practical approach to packaging and re-distributing environments in a way that attempts to improve the robustness and reproducibility of experiment results. To illustrate the versatility of our use of Unity compared to other solutions, we highlight environments already created using our approach from published papers. We hope that others can draw inspiration from how we adapted Unity to our needs, and anticipate increasingly varied and complex environments to emerge from our approach as familiarity grows.

The rapid progress in artificial intelligence (AI) and machine learning has opened unprecedented analytics possibilities in various team and individual sports, including baseball, basketball, and tennis. More recently, AI techniques have been applied to football, due to a huge increase in data collection by professional teams, increased computational power, and advances in machine learning, with the goal of better addressing new scientific challenges involved in the analysis of both individual players' and coordinated teams' behaviors. The research challenges associated with predictive and prescriptive football analytics require new developments and progress at the intersection of statistical learning, game theory, and computer vision. In this paper, we provide an overarching perspective highlighting how the combination of these fields, in particular, forms a unique microcosm for AI research, while offering mutual benefits for professional teams, spectators, and broadcasters in the years to come. We illustrate that this duality makes football analytics a game changer of tremendous value, in terms of not only changing the game of football itself, but also in terms of what this domain can mean for the field of AI. We review the state-of-the-art and exemplify the types of analysis enabled by combining the aforementioned fields, including illustrative examples of counterfactual analysis using predictive models, and the combination of game-theoretic analysis of penalty kicks with statistical learning of player attributes. We conclude by highlighting envisioned downstream impacts, including possibilities for extensions to other sports (real and virtual).

Tampering problems, where an AI agent interferes with whatever represents or communicates its intended objective and pursues the resulting corrupted objective instead, are a staple concern in the AGI safety literature [Amodei et al., 2016, Bostrom, 2014, Everitt and Hutter, 2016, Everitt et al., 2017, Armstrong and O'Rourke, 2017, Everitt and Hutter, 2019, Armstrong et al., 2020]. Variations on the idea of tampering include wireheading, where an agent learns how to stimulate its reward mechanism directly, and the off-switch or shutdown problem, where an agent interferes with its supervisor's ability to halt the agent's operation. Many real-world concerns can be formulated as tampering problems, as we will show (§2.1, §4.1). However, what constitutes tampering can be tricky to define precisely, despite clear intuitions in specific cases. We have developed a platform, REALab, to model tampering problems.

In this special guest feature, Betsy Hilliard, Principal Scientist at Valkyrie, offers three emerging trends showing how AI will play a major role in a post-COVID world and shape the business landscape moving forward. Valkyrie is a science-driven consulting firm that aims to solve organizational and global challenges through AI and machine learning. Previously, she worked as a Data Scientist at USAA Bank and RIIPL (a Rhode Island policy research lab), and interned at Google, Yahoo! and Oak Ridge National Laboratory. Betsy has a Master's in Computer Science from Brown University where she was a research assistant and published work on multi-agent learning in economic markets and Reinforcement Learning for human-agent collaboration. She studied Computer Science and Economics at Brandies University.

Machine learning (ML) algorithms have proved to be highly valuable computational tools for tackling a variety of real-world problems, including image, audio and text classification tasks. Computer scientists worldwide are developing more of these algorithms every day; thus, keeping track of them and quickly finding or accessing those introduced in the past is becoming increasingly challenging. With this in mind, researchers at Purdue University and University of Cincinnati recently created HAMLET, a platform that could help computer scientists and developers to browse through existing machine learning models and train or evaluate their own algorithms, thus aiding their research and development efforts. This platform, presented in a paper pre-published on arXiv, could ultimately democratize machine learning models developed around the world, allowing research teams to share their models with each other. "Organizing and keeping track of the machine learning algorithms and datasets has always been a major challenge for us, as well for as many other researchers in the field," Ahmad Esmaeili, one of the researchers who carried out the study, told TechXplore.

When it comes to large-scale multi-agent systems with a diverse set of agents, traditional differential privacy (DP) mechanisms are ill-matched because they consider a very broad class of adversaries, and they protect all users, independent of their characteristics, by the same guarantee. Achieving a meaningful privacy leads to pronounced reduction in solution quality. Such assumptions are unnecessary in many real-world applications for three key reasons: (i) users might be willing to disclose less sensitive information (e.g., city of residence, but not exact location), (ii) the attacker might posses auxiliary information (e.g., city of residence in a mobility-on-demand system, or reviewer expertise in a paper assignment problem), and (iii) domain characteristics might exclude a subset of solutions (an expert on auctions would not be assigned to review a robotics paper, thus there is no need for indistinguishably between reviewers on different fields). We introduce Piecewise Local Differential Privacy (PLDP), a privacy model designed to protect the utility function in applications where the attacker possesses additional information on the characteristics of the utility space. PLDP enables a high degree of privacy, while being applicable to real-world, unboundedly large settings. Moreover, we propose PALMA, a privacy-preserving heuristic for maximum-weight matching. We evaluate PALMA in a vehicle-passenger matching scenario using real data and demonstrate that it provides strong privacy, $\varepsilon \leq 3$ and a median of $\varepsilon = 0.44$, and high quality matchings ($10.8\%$ worse than the non-private optimal).

This paper presents a time-efficient scheme for Mars exploration by the cooperation of multiple drones and a rover. To maximize effective coverage of the Mars surface in the long run, a comprehensive framework has been developed with joint consideration for limited energy, sensor model, communication range and safety radius, which we call TIME-SC2 (TIme-efficient Mars Exploration of Simultaneous Coverage and Charging). First, we propose a multi-drone coverage control algorithm by leveraging emerging deep reinforcement learning and design a novel information map to represent dynamic system states. Second, we propose a near-optimal charging scheduling algorithm to navigate each drone to an individual charging slot, and we have proven that there always exists feasible solutions. The attractiveness of this framework not only resides on its ability to maximize exploration efficiency, but also on its high autonomy that has greatly reduced the non-exploring time. Extensive simulations have been conducted to demonstrate the remarkable performance of TIME-SC2 in terms of time-efficiency, adaptivity and flexibility.

For example, For many systems of interest in physics or engineering, in the case of collective particle motion, a natural choice we are given a fine-scale description of the system evolution, for such an independent variable would be the coordinates e.g. at the particle-based or agent-based level; yet the system of the space in which the particles move, and the coarsegrained exhibits large-scale, coarse-grained, spatiotemporal patterns PDE would involve the spatial derivatives of some which may well be captured by a set of unknown effective, unknown, coarse dependent variables. We assume that these coarse-grained possibly emergent PDEs. Such reduced, effective unknown dependent variables capture the local collective PDEs, when they exist and can be derived (whether (possibly averaged) statistical features of the particles, and mathematically, or in a data-driven fashion) can serve as hence can be written in terms of the local particle distribution cheap surrogate models, drastically facilitating computationintensive observations. We use manifold learning to extract tasks like prediction, optimization, uncertainty these coarse nonlinear observables from mining local particle quantification and even control.