While such systems have been shown to successfully support a broad range of interactions, they rely heavily on precoded data. For example, dialogue responses are typically limited to only one or two dozen phrases, which pales in comparison to the diversity of human speech. We believe that in order for robotic systems to become a truly ubiquitous technology, robots must make sense of natural human behavior and engage with humans in a more humanlike way. Robots must become more like humans instead of forcing humans to be more like robots. Much of human knowledge about the appropriateness of behavior, in terms of both speech and actions, comes from our personal experiences and our observations of others. We compare its performance variations form a knowledge base from which to a teleoperated robot following a scripted task we learn what to say and what actions to perform to protocol and examine both the behavior of the achieve certain goals.

Increasingly, optimization problems in machine learning, especially those arising from high-dimensional statistical estimation, have a large number of variables. Modern statistical estimators developed over the past decade have statistical or sample complexity that depends only weakly on the number of parameters when there is some structure to the problem, such as sparsity. A central question is whether similar advances can be made in their computational complexity as well. In this paper, we propose strategies that indicate that such advances can indeed be made. In particular, we investigate the greedy coordinate descent algorithm, and note that performing the greedy step efficiently weakens the costly dependence on the problem size provided the solution is sparse. We then propose a suite of methods that perform these greedy steps efficiently by a reduction to nearest neighbor search. We also devise a more amenable form of greedy descent for composite non-smooth objectives; as well as several approximate variants of such greedy descent. We develop a practical implementation of our algorithm that combines greedy coordinate descent with locality sensitive hashing. Without tuning the latter data structure, we are not only able to significantly speed up the vanilla greedy method, but also outperform cyclic descent when the problem size becomes large. Our results indicate the effectiveness of our nearest neighbor strategies, and also point to many open questions regarding the development of computational geometric techniques tailored towards first-order optimization methods.

We consider feature selection and weighting for nearest neighbor classifiers. A technical challenge in this scenario is how to cope with the discrete update of nearest neighbors when the feature space metric is changed during the learning process. This issue, called the target neighbor change, was not properly addressed in the existing feature weighting and metric learning literature. In this paper, we propose a novel feature weighting algorithm that can exactly and efficiently keep track of the correct target neighbors via sequential quadratic programming. To the best of our knowledge, this is the first algorithm that guarantees the consistency between target neighbors and the feature space metric. We further show that the proposed algorithm can be naturally combined with regularization path tracking, allowing computationally efficient selection of the regularization parameter. We demonstrate the effectiveness of the proposed algorithm through experiments.

Let us envision a new class of IT systems, the "Support Systems for Knowledge Works" or SSKW. An SSKW can be defined as a system built for providing comprehensive support to human knowledge-workers while performing instances of complex knowledge-works of a particular type within a particular domain of professional activities. To get an idea what an SSKW-enabled work environment can be like, let us look into a hypothetical scenario that depicts the interaction between a physician and a patient-care SSKW during the activity of diagnosing a patient. The patient-care task is practiced by healthcare professionals, typically within organizational setups like hospitals. An instance of the task, known as a case, is carried out by a group of professionals (physicians, surgeons, nurses, laboratory technicians etc.) led by a physician (often known as the lead physician for the case) with the primary goal of restoring an ailing patient to state of health. However, the performance also serves various secondary goals achieved through capture and reuse of information about the case. The overall task is usually divided into subtasks or activities such as examination, identification of possible diseases, clinical tests, diagnosis, treatment, followup etc. The actions taken during these activities and their results have complex interrelationships. The patient-care SSKW realizes an integrated ITbased system platform which supports all the constituent activities in ways consistent with their interrelationships. Our hypothetical scenario depicts a particular activity by the lead physician (shall be referred as LP hereafter), i.e., diagnosing a patient P with the help of a patient-care SSKW. Making a diagnosis results in identifying a particular disease based on available evidence (e.g., symptoms, signs and medical history of the patient, results of various clinical tests conducted) for which the patient will be treated. Such a scenario is described below.

Video games are complex simulation environments with many real-world properties that need to be addressed in order to build robust intelligence. In particular, real-time strategy games provide a multi-scale challenge which requires both deliberative and reactive reasoning processes. Experts approach this task by studying a corpus of games, building models for anticipating opponent actions, and practicing within the game environment. We motivate the need for integrating heterogeneous approaches by enumerating a range of competencies involved in gameplay and discuss how they are being implemented in EISBot, a reactive planning agent that we have applied to the task of playing real-time strategy games at the same granularity as humans.

Recommender systems are tools for interacting with large and complex information spaces. The field, christened in 1995, has grown enormously in the variety of problems addressed and techniques employed, as well as in its practical applications. Recommender systems research has incorporated a wide variety of artificial intelligence techniques including machine learning, data mining, user modeling, case-based reasoning, and constraint satisfaction, among others. The purpose of the articles in this special issue is to take stock of the current landscape of recommender systems research and identify directions the field is now taking.

State of the art recommender systems support users in the selection of items from a predefined assortment (for example, movies, books, and songs). In contrast to an explicit definition of each individual item, configurable products such as computers, financial service portfolios, and cars are repre¬sented in the form of a configuration knowledge base that describes the properties of allowed instances. Although the knowledge representation used is different compared to non-confi¬gurable products, the decision support requirements remain the same: users have to be supported in finding a solution that fits their wishes and needs. In this article we show how recommendation technologies can be applied for supporting the configuration of products. In addition to existing approaches we discuss relevant issues for future research.

Drama Managers, a specific type of the more general Experience Manager, have become a common subject of study in the interactive narrative literature. With a range of representational and computational approaches, authors have repeatedly developed techniques that enable computers to generate, reason about, and adapt narratives in an interactive virtual setting. In order to fully realize an experience manager, seven representational and computational problems need to be solved, generally on a case-by-case basis. In other words, the choice to use an Experience Manager is the choice to model the design as, and implement solutions to, seven inter-dependent design problems. We explicitly articulate those design problems and provide a number of examples of methods that both motivate the design problems as well as illustrate a range of approaches to solving them.

This work deals with the problem of classifying uncertain data. With this aim the Uncertain Nearest Neighbor (UNN) rule is here introduced, which represents the generalization of the deterministic nearest neighbor rule to the case in which uncertain objects are available. The UNN rule relies on the concept of nearest neighbor class, rather than on that of nearest neighbor object. The nearest neighbor class of a test object is the class that maximizes the probability of providing its nearest neighbor. It is provided evidence that the former concept is much more powerful than the latter one in the presence of uncertainty, in that it correctly models the right semantics of the nearest neighbor decision rule when applied to the uncertain scenario. An effective and efficient algorithm to perform uncertain nearest neighbor classification of a generic (un)certain test object is designed, based on properties that greatly reduce the temporal cost associated with nearest neighbor class probability computation. Experimental results are presented, showing that the UNN rule is effective and efficient in classifying uncertain data.

Diabetes management is a difficult task for patients, who must monitor and control their blood glucose levels in order to avoid serious diabetic complications. It is a difficult task for physicians, who must manually interpret large volumes of blood glucose data to tailor therapy to the needs of each patient. This paper describes three emerging applications that employ AI to ease this task and shares difficulties encountered in transitioning AI technology from university researchers to patients and physicians.