Kernel methods have been applied successfully in many applications. The kernel matrix plays an important role in kernel-based learning methods, but the ideal kernel matrix is usually unknown in practice and needs to be estimated. In this paper, we propose to directly learn the ideal kernel matrix (called the optimal neighborhood kernel matrix) from a pre-specified kernel matrix for improved classification performance. We assume that the pre-specified kernel matrix generated from the specific application is a noisy observation of the ideal one. The resulting optimal neighborhood kernel matrix is shown to be the summation of the pre-specified kernel matrix and a rank-one matrix. We formulate the problem of learning the optimal neighborhood kernel as a constrained quartic problem, and propose to solve it using two methods: level method and constrained gradient descent. Empirical results on several benchmark data sets demonstrate the efficiency and effectiveness of the proposed algorithms.

In Bayesian networks, a most probable explanation (MPE) is a most likely instantiation of all network variables given a piece of evidence. Solving (the decision version of) an MPE query is NP-hard. Recent work proposed a branch-and-bound search algorithm that finds exact solutions to MPE queries, where bounds are computed on a relaxed network obtained by a technique known as node splitting. In this work we study the impact of variable and value ordering on such a search algorithm. We study several heuristics based on the entropies of variables and on the notion of  nogoods, and propose a new meta-heuristic that combines their strengths. Experiments indicate that search efficiency is significantly improved, allowing many hard problems to be solved for the first time.

The recognition of the goal a user is pursing when interacting with a software application is a crucial task for an interface agent as it serves as a context for making opportune interventions to provide assistance to the user. The prediction of the user goal must be fast and a goal recognizer must be able to make early predictions with few observations of the user actions. In this work we propose an approach to automatically build an intention model from a plan corpus using Variable Order Markov models. We claim that following our approach, an interface agent will be capable of accurately ranking the most probable user goals in a time linear to the number of goals modeled.

In social choice, a preference function (PF) takes a set of votes (linear orders over a set of alternatives) as input, and produces one or more rankings (also linear orders over the alternatives) as output. Such functions have many applications, for example, aggregating the preferences of multiple agents, or merging rankings (of, say, webpages) into a single ranking. The key issue is choosing a PF to use. One natural and previously studied approach is to assume that there is an unobserved "correct" ranking, and the votes are noisy estimates of this. Then, we can use the PF that always chooses the maximum likelihood estimate (MLE) of the correct ranking. In this paper, we define simple ranking scoring functions (SRSFs) and show that the class of neutral SRSFs is exactly the class of neutral PFs that are MLEs for some noise model. We also define composite ranking scoring functions (CRSFs) and show a condition under which these coincide with SRSFs. We study key properties such as consistency and continuity, and consider some example PFs. In particular, we study Single Transferable Vote (STV), a commonly used PF, showing that it is a CRSF but not an SRSF, thereby clarifying the extent to which it is an MLE function. This also gives a new perspective on how ties should be broken under STV. We leave some open questions.

Linear causal models known as structural equation models (SEMs) are widely used for data analysis in the social sciences, economics, and artificial intelligence, in which random variables are assumed to be continuous and normally distributed. This paper deals with one fundamental problem in the applications of SEMs -- parameter identification. The paper uses the graphical models approach and provides a procedure for solving the identification problem in a special class of SEMs.

Robots operating in home environments must be able to interact with articulated objects such as doors or drawers.  Ideally, robots are able to autonomously infer articulation models by observation.  In this paper, we present an approach to learn kinematic models by inferring the connectivity of rigid parts and the articulation models for the corresponding links.  Our method uses a mixture of parameterized and parameter-free (Gaussian process) representations and finds low-dimensional manifolds that provide the best explanation of the given observations.  Our approach has been implemented and evaluated using real data obtained in various realistic home environment settings.

Sensors provide computer systems with a window to the outside world. Activity recognition "sees" what is in the window to predict the locations, trajectories, actions, goals and plans of humans and objects. Building an activity recognition system requires a full range of interaction from statistical inference on lower level sensor data to symbolic AI at higher levels, where prediction results and acquired knowledge are passed up each level to form a knowledge food chain. In this article, I will give an overview of some of the current activity recognition research works and explore a life-cycle of learning and inference that allows the lowest-level radio-frequency signals to be transformed into symbolic logical representations for AI planning, which in turn controls the robots or guides human users through a sensor network, thus completing a full life-cycle of knowledge.

On the other hand, the Recently, Ferraris, Lee and Lifschitz presented a solution provided by answer set programming (ASP), that is general definition of a stable model that is similar carried over to high level action language A [Gelfond and to the definition of circumscription, and can even Lifschitz, 1998] and many of its descendants that are based be characterized in terms of circumscription. In on ASP, uses both default negation (not) and strong negation this paper, we show the opposite direction, which ()--the idea of which is closely related to Reiter's default is, how to turn circumscription into the general stable logic solution [Reiter, 1980]. Interestingly, the development model semantics, and based on this, how to turn of the event calculus has spanned over both classical circumscriptive event calculus into answer set programs.

Qualitative constraint calculi are representation formalisms that allow for efficient reasoning about spatial and temporal information. Many of the calculi discussed in the field of Qualitative Spatial and Temporal Reasoning can be defined as combinations of other, simpler and more compact formalisms. On the other hand, existing calculi can be combined to a new formalism in which one can represent, and reason about, different aspects of a domain at the same time. For example, Gerevini and Renz presented a loose combination of the region connection calculus RCC-8 and the point algebra: the resulting formalism integrates topological and qualitative size relations between spatially extended objects. In this paper we compare the approach by Gerevini and Renz to a method that generates a new qualitative calculus by exploiting the semantic interdependencies between the component calculi. We will compare these two methods and analyze some formal relationships between a combined calculus and its components. The paper is completed by an empirical case study in which the reasoning performance of the suggested methods is compared on random test instances.

Early research in heuristic search discovered that using inconsistent heuristics with A* could result in an exponential increase in the number of node expansions. As a result, the use of inconsistent heuristics has largely disappeared from practice. Recently, inconsistent heuristics have been shown to be effective in IDA*, especially when applying the bidirectional pathmax (BPMX) enhancement. This paper presents new worst-case complexity analysis of A*'s behavior with inconsistent heuristics, discusses how BPMX can be used with A*, and gives experimental results justifying the use of inconsistent heuristics in A* searches.