For nearly 25 years my research group has investigated the use of domain knowledge, expressed in some version of mathematical logic, that is refined or exploited by numeric-based learning algorithms. These include what we called knowledge-based neural networks and knowledge-based support vector machines. I will cover the key ideas of these methods, as well as the behind-the-scenes motivations that lead to them. I will also describe why we switched from using the phrase 'prior knowledge' to using 'advice.' Finally, I will cover some of our recent work on fast learning and inference for Markov Logic Networks (which can be viewed as a knowledge-based graphical model).

Optimal solutions for multi-agent pathfinding problems are often too expensive to compute. For this reason, suboptimal approaches have been widely studied in the literature. Specifically, in recent years a number of efficient suboptimal algorithms that are complete for certain subclasses have been proposed at highly-rated robotics and AI conferences. However, it turns out that the problem of non-optimal multi-agent pathfinding has already been completely solved in another research community in the 1980s. In this paper, we would like to bring this earlier related work to the attention of the robotics and AI communities.

Several variants of A* have been recently proposed for find-ing optimal solutions for the multi-agent pathfinding (MAPF)problem. However, these variants have not been deeply com-pared either quantitatively or qualitatively. In this paper weaim to fill this gap. In addition to obtaining a deeper under-standing of the existing algorithms, we describe in detail theapplication of the new enhanced partial-expansion techniqueto MAPF and show how pattern databases can be applied ontop of this technique.

In our ongoing work, we aim to control a team of agents soas to achieve a prescribed goal state while being confidentthat collisions with other agents are avoided. Each agent isassociated with a feedback controlled plant, whose continu-ous state trajectories follow some stochastic differential dy-namics. To this end we describe a collision-detection modulebased on a distribution-independent probabilistic bound andemploy a fixed priority method to resolve collisions. Dueto their practical importance, multi-agent collision avoid-ance and control have been extensively studied across differ-ent communities including AI, robotics and control. How-ever, these works typically assume linear and discrete dy-namic models; by contrast, our work intends to overcomethese limitations and to present solutions for continuousstate space. While our current experiments were conductedwith linear stochastic differential equation (SDE) modelswith state-independent noise (yielding Gaussian processes)we believe that our approach could also be applicable to non-Gaussian cases with state-dependent uncertainties.

We present a new weighted session similarity measure to capture the browsing interests of users in web usage profiles discovered from web log data. We base our similarity measure on the reasonable assumption that when users spend longer times on pages or revisit pages in the same session, then very likely, such pages are of greater interest to the user. The proposed similarity measure combines structural similarity with session-wise page significance. The latter, representing the degree of user interest, is computed using frequency and duration of a page access. Web usage profiles are generated using this similarity measure by applying a fuzzy clustering algorithm to web log data. For evaluating the effectiveness of the proposed measure, we adapt two model-based collaborative filtering algorithms for recommending pages. Experimental results show considerable improvement in overall performance of recommender systems as compared to use of other existing similarity measures.

Computational systems can use sociocultural knowledge to understand human behavior and interact with humans in more natural ways. However, such systems are limited by their reliance on hand-authored sociocultural knowledge and models. We introduce an approach to automatically learn robust, script-like sociocultural knowledge from crowdsourced narratives. Crowdsourcing, the use of anonymous human workers, provides an opportunity for rapidly acquir­ing a corpus of examples of situations that are highly specialized for our purpose yet sufficiently varied, from which we can learn a versatile script. We describe a semi-automated process by which we query human workers to write natural language narrative examples of a given situation and learn the set of events that can occur and the typical even ordering.

Over a series of evaluation experiments conducted using naive judges recruited and managed via Amazon's Mechanical Turk facility using a task from information retrieval (IR), we show that a SVM shows itself to have a very high accuracy when the machine-learner is trained and tested on a single task and that the method was portable from more complex tasks to simpler tasks, but not vice versa.

In crowdsourced relevance judging, each crowd workertypically judges only a small number of examples,yielding a sparse and imbalanced set of judgments inwhich relatively few workers influence output consensuslabels, particularly with simple consensus methodslike majority voting. We show how probabilistic matrixfactorization, a standard approach in collaborative filtering,can be used to infer missing worker judgments suchthat all workers influence output labels. Given completeworker judgments inferred by PMF, we evaluate impactin unsupervised and supervised scenarios. In thesupervised case, we consider both weighted voting andworker selection strategies based on worker accuracy.Experiments on a synthetic data set and a real turk dataset with crowd judgments from the 2010 TREC RelevanceFeedback Track show promise of the PMF approachmerits further investigation and analysis.

Crowdsourcing markets, such as Amazon Mechanical Turk, provide a platform for matching prospective workers around the world with tasks. However, they are often plagued by workers who attempt to exert as little effort as possible, and requesters who deny workers payment for their labor. For crowdsourcing markets to succeed, it is essential to discourage such behavior. With this in mind, we propose a framework for the design and analysis of incentive mechanisms based on social norms, which consist of a set of rules that participants are expected to follow, and a mechanism for updating participants’ public reputations based on whether or not they do. We start by considering the most basic version of our model, which contains only homogeneous participants and randomly matches workers with tasks. The optimal social norm in this setting turns out to be a simple, easily comprehensible incentive mechanism in which market participants are encouraged to play a tit-for-tat-like strategy. This simple mechanism is optimal even when the set of market participants changes dynamically over time, or when some fraction of the participants may be irrational. In addition to the basic model, we demonstrate how this framework can be applied to situations in which there are heterogeneous users by giving several illustrating examples. This work is a first step towards a complete theory of incentive design for crowdsourcing systems. We hope to build upon this framework and explore more interesting and practical aspects of real online labor markets in our future work.

We explore the use of an interface to mark pairs of points on two images which are in "correspondence" with one another, as a way of collecting part annotations. The interface allows annotations of visual categories that are structurally diverse, such as chairs and buildings, where it is difficult to define a set of parts, or landmarks, that are consistent, namable or uniquely defined across all instances of the category. It allows flexibility in annotation - the landmarks can be instance specific, are not constrained by language, could be many to one, etc and requires little category specific instructions. We compare our approach to two popular methods of collecting part annotations, (1) drawing bounding boxes for a set of parts, and (2) annotating a set of landmarks, in terms of annotation setup overhead, cost, difficulty, applicability and utility, and identify scenarios where one method is better suited than the others. Preliminary experiments suggest that such annotations between a sparse set of pairs can be used to bootstrap many high level visual recognition tasks such as part discovery and semantic saliency.