Our research explored a new approach to an old problem: we reformulated principal component analysis (PCA), a type of eigenvalue problem, as a competitive multi-agent game we call EigenGame.

A key challenge on the path to developing agents that learn complex human-like behavior is the need to quickly and accurately quantify human-likeness. While human assessments of such behavior can be highly accurate, speed and scalability are limited. We address these limitations through a novel automated Navigation Turing Test (ANTT) that learns to predict human judgments of human-likeness. We demonstrate the effectiveness of our automated NTT on a navigation task in a complex 3D environment. We investigate six classification models to shed light on the types of architectures best suited to this task, and validate them against data collected through a human NTT. Our best models achieve high accuracy when distinguishing true human and agent behavior. At the same time, we show that predicting finer-grained human assessment of agents' progress towards human-like behavior remains unsolved. Our work takes an important step towards agents that more effectively learn complex human-like behavior.

Abstract-- In recent years, we have witnessed increasingly high performance in the field of autonomous end-toend driving. In particular, more and more research is being done on driving in urban environments, where the car has to follow high level commands to navigate. However, few evaluations are made on the ability of these agents to react in an unexpected situation. Specifically, no evaluations are conducted on the robustness of driving agents in the event of a bad high-level command. We propose here an evaluation method, namely a benchmark that allows to assess the robustness of an agent, and to appreciate its understanding of the environment through its ability to keep a safe behavior, regardless of the instruction.

Where information grows abundant, attention becomes a scarce resource. As a result, agents must plan wisely how to allocate their attention in order to achieve epistemic efficiency. Here, we present a framework for multi-agent epistemic planning with attention, based on Dynamic Epistemic Logic (DEL, a powerful formalism for epistemic planning). We identify the framework as a fragment of standard DEL, and consider its plan existence problem. While in the general case undecidable, we show that when attention is required for learning, all instances of the problem are decidable.

This study reviews research on social emotions in robotics. In robotics, emotions are pursued for a long duration, such as recognition, expression, and computational modeling of the basic mechanism behind them. Research has been promoted according to well-known psychological findings, such as category and dimension theories. Many studies have been based on these basic theories, addressing only basic emotions. However, social emotions, also called higher-level emotions, have been studied in psychology. We believe that these higher-level emotions are worth pursuing in robotics for next-generation social-aware robots. In this review paper, while summarizing the findings of social emotions in psychology and neuroscience, studies on social emotions in robotics at present are surveyed. Thereafter, research directions towards implementation of social emotions in robots are discussed.

The maximin share (MMS) guarantee is a desirable fairness notion for allocating indivisible goods. While MMS allocations do not always exist, several approximation techniques have been developed to ensure that all agents receive a fraction of their maximin share. We focus on an alternative approximation notion, based on the population of agents, that seeks to guarantee MMS for a fraction of agents. We show that no optimal approximation algorithm can satisfy more than a constant number of agents, and discuss the existence and computation of MMS for all but one agent and its relation to approximate MMS guarantees. We then prove the existence of allocations that guarantee MMS for $\frac{2}{3}$ of agents, and devise a polynomial time algorithm that achieves this bound for up to nine agents. A key implication of our result is the existence of allocations that guarantee $\text{MMS}^{\lceil{3n/2}\rceil}$, i.e., the value that agents receive by partitioning the goods into $\lceil{\frac{3}{2}n}\rceil$ bundles, improving the best known guarantee of $\text{MMS}^{2n-2}$. Finally, we provide empirical experiments using synthetic data.

As AI systems are increasingly involved in decision making, it also becomes important that they elicit appropriate levels of trust from their users. To achieve this, it is first important to understand which factors influence trust in AI. We identify that a research gap exists regarding the role of personal values in trust in AI. Therefore, this paper studies how human and agent Value Similarity (VS) influences a human's trust in that agent. To explore this, 89 participants teamed up with five different agents, which were designed with varying levels of value similarity to that of the participants. In a within-subjects, scenario-based experiment, agents gave suggestions on what to do when entering the building to save a hostage. We analyzed the agent's scores on subjective value similarity, trust and qualitative data from open-ended questions. Our results show that agents rated as having more similar values also scored higher on trust, indicating a positive effect between the two. With this result, we add to the existing understanding of human-agent trust by providing insight into the role of value-similarity.

I describe the rationale for, and design of, an agent-based simulation model of a contemporary online sports-betting exchange: such exchanges, closely related to the exchange mechanisms at the heart of major financial markets, have revolutionized the gambling industry in the past 20 years, but gathering sufficiently large quantities of rich and temporally high-resolution data from real exchanges - i.e., the sort of data that is needed in large quantities for Deep Learning - is often very expensive, and sometimes simply impossible; this creates a need for a plausibly realistic synthetic data generator, which is what this simulation now provides. The simulator, named the "Bristol Betting Exchange" (BBE), is intended as a common platform, a data-source and experimental test-bed, for researchers studying the application of AI and machine learning (ML) techniques to issues arising in betting exchanges; and, as far as I have been able to determine, BBE is the first of its kind: a free open-source agent-based simulation model consisting not only of a sports-betting exchange, but also a minimal simulation model of racetrack sporting events (e.g., horse-races or car-races) about which bets may be made, and a population of simulated bettors who each form their own private evaluation of odds and place bets on the exchange before and - crucially - during the race itself (i.e., so-called "in-play" betting) and whose betting opinions change second-by-second as each race event unfolds. BBE is offered as a proof-of-concept system that enables the generation of large high-resolution data-sets for automated discovery or improvement of profitable strategies for betting on sporting events via the application of AI/ML and advanced data analytics techniques. This paper offers an extensive survey of relevant literature and explains the motivation and design of BBE, and presents brief illustrative results.

In many real-world scenarios, a team of agents coordinate with each other to compete against an opponent. The challenge of solving this type of game is that the team's joint action space grows exponentially with the number of agents, which results in the inefficiency of the existing algorithms, e.g., Counterfactual Regret Minimization (CFR). To address this problem, we propose a new framework of CFR: CFR-MIX. Firstly, we propose a new strategy representation that represents a joint action strategy using individual strategies of all agents and a consistency relationship to maintain the cooperation between agents. To compute the equilibrium with individual strategies under the CFR framework, we transform the consistency relationship between strategies to the consistency relationship between the cumulative regret values. Furthermore, we propose a novel decomposition method over cumulative regret values to guarantee the consistency relationship between the cumulative regret values. Finally, we introduce our new algorithm CFR-MIX which employs a mixing layer to estimate cumulative regret values of joint actions as a non-linear combination of cumulative regret values of individual actions. Experimental results show that CFR-MIX outperforms existing algorithms on various games significantly.

In robotics, ergodic control extends the tracking principle by specifying a probability distribution over an area to cover instead of a trajectory to track. The original problem is formulated as a spectral multiscale coverage problem, typically requiring the spatial distribution to be decomposed as Fourier series. This approach does not scale well to control problems requiring exploration in search space of more than 2 dimensions. To address this issue, we propose the use of tensor trains, a recent low-rank tensor decomposition technique from the field of multilinear algebra. The proposed solution is efficient, both computationally and storage-wise, hence making it suitable for its online implementation in robotic systems. The approach is applied to a peg-in-hole insertion task requiring full 6D end-effector poses, implemented with a 7-axis Franka Emika Panda robot. In this experiment, ergodic exploration allows the task to be achieved without requiring the use of force/torque sensors.