Standard Epistemic Logic (EL) mainly studies reasoning patterns of knowing that ϕ, despite early contributions by Hintikka on formulating other know-wh expressions such as knowing who and why using first-order and higher-order modal logic. In recent years, there is a resurgence of interest on epistemic logics of know-wh powered by the new techniques for fragments of firstorder modal logic based on the so-called bundle modalities packing a quantifier and a normal epistemic modality together [26, 24, 21]. Within the varieties of logics of know-wh, the logics of know-how received the most attention in AI (cf.

Rethinking cities is now more imperative than ever, as society faces global challenges such as population growth and climate change. The design of cities can not be abstracted from the design of its mobility system, and, therefore, efficient solutions must be found to transport people and goods throughout the city in an ecological way. An autonomous bicycle-sharing system would combine the most relevant benefits of vehicle sharing, electrification, autonomy, and micro-mobility, increasing the efficiency and convenience of bicycle-sharing systems and incentivizing more people to bike and enjoy their cities in an environmentally friendly way. Due to the uniqueness and radical novelty of introducing autonomous driving technology into bicycle-sharing systems and the inherent complexity of these systems, there is a need to quantify the potential impact of autonomy on fleet performance and user experience. This paper presents an ad-hoc agent-based simulator that provides an in-depth understanding of the fleet behavior of autonomous bicycle-sharing systems in realistic scenarios, including a rebalancing system based on demand prediction. In addition, this work describes the impact of different parameters on system efficiency and service quality and quantifies the extent to which an autonomous system would outperform current bicycle-sharing schemes. The obtained results show that with a fleet size three and a half times smaller than a station-based system and eight times smaller than a dockless system, an autonomous system can provide overall improved performance and user experience even with no rebalancing. These findings indicate that the remarkable efficiency of an autonomous bicycle-sharing system could compensate for the additional cost of autonomous bicycles.

We investigate Kantian equilibria in finite normal form games, a class of non-Nashian, morally motivated courses of action that was recently proposed in the economics literature. We highlight a number of problems with such equilibria, including computational intractability, a high price of miscoordination, and problematic extension to general normal form games. We give such a generalization based on concept of program equilibria, and point out that that a practically relevant generalization may not exist. To remedy this we propose some general, intuitive, computationally tractable, other-regarding equilibria that are special cases Kantian equilibria, as well as a class of courses of action that interpolates between purely self-regarding and Kantian behavior.

We study online Multi-Agent Path Finding (MAPF), where new agents are constantly revealed over time and all agents must find collision-free paths to their given goal locations. We generalize existing complexity results of (offline) MAPF to online MAPF. We classify online MAPF algorithms into different categories based on (1) controllability (the set of agents that they can plan paths for at each time) and (2) rationality (the quality of paths they plan) and study the relationships between them. We perform a competitive analysis for each category of online MAPF algorithms with respect to commonly-used objective functions. We show that a naive algorithm that routes newly-revealed agents one at a time in sequence achieves a competitive ratio that is asymptotically bounded from both below and above by the number of agents with respect to flowtime and makespan. We then show a counter-intuitive result that, if rerouting of previously-revealed agents is not allowed, any rational online MAPF algorithms, including ones that plan optimal paths for all newly-revealed agents, have the same asymptotic competitive ratio as the naive algorithm, even on 2D 4-neighbor grids. We also derive constant lower bounds on the competitive ratio of any rational online MAPF algorithms that allow rerouting. The results thus provide theoretical insights into the effectiveness of using MAPF algorithms in an online setting for the first time.

Multi-Agent Path Finding (MAPF) is essential to large-scale robotic systems. Recent methods have applied reinforcement learning (RL) to learn decentralized polices in partially observable environments. A fundamental challenge of obtaining collision-free policy is that agents need to learn cooperation to handle congested situations. This paper combines communication with deep Q-learning to provide a novel learning based method for MAPF, where agents achieve cooperation via graph convolution. To guide RL algorithm on long-horizon goal-oriented tasks, we embed the potential choices of shortest paths from single source as heuristic guidance instead of using a specific path as in most existing works. Our method treats each agent independently and trains the model from a single agent's perspective. The final trained policy is applied to each agent for decentralized execution. The whole system is distributed during training and is trained under a curriculum learning strategy. Empirical evaluation in obstacle-rich environment indicates the high success rate with low average step of our method.

Involving humans directly for the benefit of AI agents' training is getting traction thanks to several advances in reinforcement learning and human-in-the-loop learning. Humans can provide rewards to the agent, demonstrate tasks, design a curriculum, or act in the environment, but these benefits also come with architectural, functional design and engineering complexities. We present Cogment, a unifying open-source framework that introduces an actor formalism to support a variety of humans-agents collaboration typologies and training approaches. It is also scalable out of the box thanks to a distributed micro service architecture, and offers solutions to the aforementioned complexities.

We propose a curriculum-driven learning strategy for solving difficult multi-agent coordination tasks. Our method is inspired by a study of animal communication, which shows that two straightforward design features (mutual reward and decentralization) support a vast spectrum of communication protocols in nature. We highlight the importance of similarly interpreting emergent communication as a spectrum. We introduce a toroidal, continuous-space pursuit-evasion environment and show that naive decentralized learning does not perform well. We then propose a novel curriculum-driven strategy for multi-agent learning. Experiments with pursuit-evasion show that our approach enables decentralized pursuers to learn to coordinate and capture a superior evader, significantly outperforming sophisticated analytical policies. We argue through additional quantitative analysis -- including influence-based measures such as Instantaneous Coordination -- that emergent implicit communication plays a large role in enabling superior levels of coordination.

The TARK conference (Theoretical Aspects of Rationality and Knowledge) is a biannual conference that aims to bring together researchers from a wide variety of fields, including computer science, artificial intelligence, game theory, decision theory, philosophy, logic, linguistics, and cognitive science. Its goal is to further our understanding of interdisciplinary issues involving reasoning about rationality and knowledge. Topics of interest include, but are not limited to, semantic models for knowledge, belief, awareness and uncertainty, bounded rationality and resource-bounded reasoning, commonsense epistemic reasoning, epistemic logic, epistemic game theory, knowledge and action, applications of reasoning about knowledge and other mental states, belief revision, and foundations of multi-agent systems. These proceedings contain the papers that have been accepted for presentation at the Eighteenth Conference on Theoretical Aspects of Rationality and Knowledge (TARK 2021), held between June 25 and June 27, 2021, at Tsinghua University at Beijing, China.

When studying robots collaborating with humans, much of the focus has been on robot policies that coordinate fluently with human teammates in collaborative tasks. However, less emphasis has been placed on the effect of the environment on coordination behaviors. To thoroughly explore environments that result in diverse behaviors, we propose a framework for procedural generation of environments that are (1) stylistically similar to human-authored environments, (2) guaranteed to be solvable by the human-robot team, and (3) diverse with respect to coordination measures. We analyze the procedurally generated environments in the Overcooked benchmark domain via simulation and an online user study. Results show that the environments result in qualitatively different emerging behaviors and statistically significant differences in collaborative fluency metrics, even when the robot runs the same planning algorithm.

The evaluation of rare but high-stakes events remains one of the main difficulties in obtaining reliable policies from intelligent agents, especially in large or continuous state/action spaces where limited scalability enforces the use of a prohibitively large number of testing iterations. On the other hand, a biased or inaccurate policy evaluation in a safety-critical system could potentially cause unexpected catastrophic failures during deployment. In this paper, we propose the Accelerated Policy Evaluation (APE) method, which simultaneously uncovers rare events and estimates the rare event probability in Markov decision processes. The APE method treats the environment nature as an adversarial agent and learns towards, through adaptive importance sampling, the zero-variance sampling distribution for the policy evaluation. Moreover, APE is scalable to large discrete or continuous spaces by incorporating function approximators. We investigate the convergence properties of proposed algorithms under suitable regularity conditions. Our empirical studies show that APE estimates rare event probability with a smaller variance while only using orders of magnitude fewer samples compared to baseline methods in both multi-agent and single-agent environments.