Goto

Collaborating Authors

Journal of Artificial Intelligence Research


Journal of Artificial Intelligence Research

Journal of Artificial Intelligence Research

JAIR is published by AI Access Foundation, a nonprofit public charity whose purpose is to facilitate the dissemination of scientific results in artificial intelligence. JAIR, established in 1993, was one of the first open-access scientific journals on the Web, and has been a leading publication venue since its inception. We invite you to check out our other initiatives.


FAQ

Journal of Artificial Intelligence Research

JAIR publishes full-length research articles, short technical notes, and survey and expository articles. See the editorial charter for more details. JAIR distributes articlees for free on the public internet. Individual users may read, download, copy, distribute, print, search, or link to the full texts of indvidual articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose as specified in our licensing terms. However, the right to publish complete volumes of the journal in print is reserved exclusively for AAAI, and therefore others may not sell or publish complete volumes of JAIR in print.


Preferences Single-Peaked on a Circle

Journal of Artificial Intelligence Research

We introduce the domain of preferences that are single-peaked on a circle, which is a generalization of the well-studied single-peaked domain. This preference restriction is useful, e.g., for scheduling decisions, certain facility location problems, and for one-dimensional decisions in the presence of extremist preferences. We give a fast recognition algorithm of this domain, provide a characterisation by finitely many forbidden subprofiles, and show that many popular single- and multi-winner voting rules are polynomial-time computable on this domain. In particular, we prove that Proportional Approval Voting can be computed in polynomial time for profiles that are single-peaked on a circle. In contrast, Kemeny's rule remains hard to evaluate, and several impossibility results from social choice theory can be proved using only profiles in this domain.


A General Approach to Multimodal Document Quality Assessment

Journal of Artificial Intelligence Research

The perceived quality of a document is affected by various factors, including grammat- icality, readability, stylistics, and expertise depth, making the task of document quality assessment a complex one. In this paper, we explore this task in the context of assessing the quality of Wikipedia articles and academic papers. Observing that the visual rendering of a document can capture implicit quality indicators that are not present in the document text — such as images, font choices, and visual layout — we propose a joint model that combines the text content with a visual rendering of the document for document qual- ity assessment. Our joint model achieves state-of-the-art results over five datasets in two domains (Wikipedia and academic papers), which demonstrates the complementarity of textual and visual features, and the general applicability of our model. To examine what kinds of features our model has learned, we further train our model in a multi-task learning setting, where document quality assessment is the primary task and feature learning is an auxiliary task. Experimental results show that visual embeddings are better at learning structural features while textual embeddings are better at learning readability scores, which further verifies the complementarity of visual and textual features.


Simulating Offender Mobility: Modeling Activity Nodes from Large-Scale Human Activity Data

Journal of Artificial Intelligence Research

In recent years, simulation techniques have been applied to investigate the spatiotemporal dynamics of crime. Researchers have instantiated mobile offenders in agent-based simulations for theory testing, experimenting with crime prevention strategies, and exploring crime prediction techniques, despite facing challenges due to the complex dynamics of crime and the lack of detailed information about offender mobility. This paper presents a simulation model to explore offender mobility, focusing on the interplay between the agent's awareness space and activity nodes. The simulation generates patterns of individual mobility aiming to cumulatively match crime patterns. To instantiate a realistic urban environment, we use open data to simulate the urban structure, location-based social networks data to represent activity nodes as a proxy for human activity, and taxi trip data as a proxy for human movement between regions of the city. We analyze and systematically compare 35 different mobility strategies and demonstrate the benefits of using large-scale human activity data to simulate offender mobility. The strategies combining taxi trip data or historic crime data with popular activity nodes perform best compared to other strategies, especially for robbery. Our approach provides a basis for building agent-based crime simulations that infer offender mobility in urban areas from real-world data.


Predicting Strategic Behavior from Free Text

Journal of Artificial Intelligence Research

The connection between messaging and action is fundamental both to web applications, such as web search and sentiment analysis, and to economics. However, while prominent online applications exploit messaging in natural (human) language in order to predict non-strategic action selection, the economics literature focuses on the connection between structured stylized messaging to strategic decisions in games and multi-agent encounters. This paper aims to connect these two strands of research, which we consider highly timely and important due to the vast online textual communication on the web. Particularly, we introduce the following question: Can free text expressed in natural language serve for the prediction of action selection in an economic context, modeled as a game? In order to initiate the research on this question, we introduce the study of an individual's action prediction in a one-shot game based on free text he/she provides, while being unaware of the game to be played. We approach the problem by attributing commonsensical personality attributes via crowd-sourcing to free texts written by individuals, and employing transductive learning to predict actions taken by these individuals in one-shot games based on these attributes. Our approach allows us to train a single classifier that can make predictions with respect to actions taken in multiple games. In experiments with three well-studied games, our algorithm compares favorably with strong alternative approaches. In ablation analysis, we demonstrate the importance of our modeling choices--the representation of the text with the commonsensical personality attributes and our classifier--to the predictive power of our model.


The Effects of Experience on Deception in Human-Agent Negotiation

Journal of Artificial Intelligence Research

Negotiation is the complex social process by which multiple parties come to mutual agreement over a series of issues. As such, it has proven to be a key challenge problem for designing adequately social AIs that can effectively navigate this space. Artificial AI agents that are capable of negotiating must be capable of realizing policies and strategies that govern offer acceptances, offer generation, preference elicitation, and more. But the next generation of agents must also adapt to reflect their users’ experiences.      The best human negotiators tend to have honed their craft through hours of practice and experience. But, not all negotiators agree on which strategic tactics to use, and endorsement of deceptive tactics in particular is a controversial topic for many negotiators. We examine the ways in which deceptive tactics are used and endorsed in non-repeated human negotiation and show that prior experience plays a key role in governing what tactics are seen as acceptable or useful in negotiation. Previous work has indicated that people that negotiate through artificial agent representatives may be more inclined to fairness than those people that negotiate directly. We present a series of three user studies that challenge this initial assumption and expand on this picture by examining the role of past experience.      This work constructs a new scale for measuring endorsement of manipulative negotiation tactics and introduces its use to artificial intelligence research. It continues by presenting the results of a series of three studies that examine how negotiating experience can change what negotiation tactics and strategies human endorse. Study #1 looks at human endorsement of deceptive techniques based on prior negotiating experience as well as representative effects. Study #2 further characterizes the negativity of prior experience in relation to endorsement of deceptive techniques. Finally, in Study #3, we show that the lessons learned from the empirical observations in Study #1 and #2 can in fact be induced—by designing agents that provide a specific type of negative experience, human endorsement of deception can be predictably manipulated.


Sliding-Window Thompson Sampling for Non-Stationary Settings

Journal of Artificial Intelligence Research

Multi-Armed Bandit (MAB) techniques have been successfully applied to many classes of sequential decision problems in the past decades. However, non-stationary settings -- very common in real-world applications -- received little attention so far, and theoretical guarantees on the regret are known only for some frequentist algorithms. In this paper, we propose an algorithm, namely Sliding-Window Thompson Sampling (SW-TS), for nonstationary stochastic MAB settings. Our algorithm is based on Thompson Sampling and exploits a sliding-window approach to tackle, in a unified fashion, two different forms of non-stationarity studied separately so far: abruptly changing and smoothly changing. In the former, the reward distributions are constant during sequences of rounds, and their change may be arbitrary and happen at unknown rounds, while, in the latter, the reward distributions smoothly evolve over rounds according to unknown dynamics. Under mild assumptions, we provide regret upper bounds on the dynamic pseudo-regret of SW-TS for the abruptly changing environment, for the smoothly changing one, and for the setting in which both the non-stationarity forms are present. Furthermore, we empirically show that SW-TS dramatically outperforms state-of-the-art algorithms even when the forms of non-stationarity are taken separately, as previously studied in the literature.


Improving Nash Social Welfare Approximations

Journal of Artificial Intelligence Research

We consider the problem of fairly allocating a set of indivisible goods among n agents. Various fairness notions have been proposed within the rapidly growing field of fair division, but the Nash social welfare (NSW) serves as a focal point. In part, this follows from the ‘unreasonable’ fairness guarantees provided, in the sense that a max NSW allocation meets multiple other fairness metrics simultaneously, all while satisfying a standard economic concept of efficiency, Pareto optimality. However, existing approximation algorithms fail to satisfy all of the remarkable fairness guarantees offered by a max NSW allocation, instead targeting only the specific NSW objective. We address this issue by presenting a 2 max NSW, Prop-1, 1/(2n) MMS, and Pareto optimal allocation in strongly polynomial time. Our techniques are based on a market interpretation of a fractional max NSW allocation. We present novel definitions of fairness concepts in terms of market prices, and design a new scheme to round a market equilibrium into an integral allocation in a way that provides most of the fairness properties of an integral max NSW allocation.


Towards Partial Order Reductions for Strategic Ability

Journal of Artificial Intelligence Research

We propose a general semantics for strategic abilities of agents in asynchronous systems, with and without perfect information. Based on the semantics, we show some general complexity results for verification of strategic abilities in asynchronous interaction. More importantly, we develop a methodology for partial order reduction in verification of agents with imperfect information. We show that the reduction preserves an important subset of strategic properties, with as well as without the fairness assumption. We also demonstrate the effectiveness of the reduction on a number of benchmarks. Interestingly, the reduction does not work for strategic abilities under perfect information.