This paper presents a new algorithm for plan recognition called ELEXIR (Engine for LEXicalized Intent Recognition).  ELEXIR represents the plans to be recognized with a grammatical formalism called Combinatory Categorial Grammar(CCG).  We show that representing plans with CCGs can allow us to prevent early commitment to plan goals and thereby reduce runtime.

Model-Based Diagnosis (MBD) approaches often yield a large number of diagnoses, severely limiting their practical utility. This paper presents a novel active testing approach based on MBD techniques, called FRACTAL (FRamework for ACtive Testing ALgorithms), which, given a system description, computes a sequence of control settings for reducing the number of diagnoses. The approach complements probing, sequential diagnosis, and ATPG, and applies to systems where additional tests are restricted to setting a subset of the existing system inputs while observing the existing outputs. This paper evaluates the optimality of FRACTAL, both theoretically and empirically. FRACTAL generates test vectors using a greedy, next-best strategy and a low-cost approximation of diagnostic information entropy. Further, the approximate sequence computed by FRACTAL's greedy approach is optimal over all poly-time approximation algorithms, a fact which we confirm empirically. Extensive experimentation with ISCAS85 combinational circuits shows that FRACTAL reduces the number of remaining diagnoses according to a steep geometric decay function, even when only a fraction of inputs are available for active testing.

In this work, we propose automatically generating a list of view filters relevant to the displayed messages. Our filters Growing email volumes cause flooded inboxes and are implemented as searches, such as a search for all messages swelled email archives, making search and new in the inbox from a discussion list. We call our task email processing difficult. While emails have rich Search Operator Suggestion, where search operators are special metadata, such as recipients and folders, suitable terms that retrieve emails based on message metadata, for creating filtered views, it is often difficult to such as "from:john smith" and "is:starred." We build a mail choose appropriate filters for new inbox messages filter system for Gmail (Google Mail) using search operators without first examining messages. In this work, we and develop several search operator rankers using features of consider a system that automatically suggests relevant the user, mailbox and machine learning. We validate our system view filters to the user for the currently viewed on data collected from user interactions with our system.

While much work on learning in planning focused on learning domain physics (i.e., action models), and search control knowledge, little attention has been paid towards learning user preferences on desirable plans. Hierarchical task networks (HTN) are known to provide an effective way to encode user prescriptions about what constitute good plans. However, manual construction of these methods is complex and error prone. In this paper, we propose a novel approach to learning probabilistic hierarchical task networks that capture user preferences by examining user-produced plans given no prior information about the methods (in contrast, most prior work on learning within the HTN framework focused on learning “method preconditions”—i.e., domain physics—assuming that the structure of the methods is given as input). We will show that this problem has close parallels to the problem of probabilistic grammar induction, and describe how grammar induction methods can be adapted to learn task networks. We will empirically demonstrate the effectiveness of our approach by showing that task networks we learn are able to generate plans with a distribution close to the distribution of the userpreferred plans.

Distance metric has an important role in many machine learning algorithms. Recently, metric learning for semi-supervised algorithms has received much attention. For semi-supervised clustering, usually a set of pairwise similarity and dissimilarity constraints is provided as supervisory information. Until now, various metric learning methods utilizing pairwise constraints have been proposed. The existing methods that can consider both positive (must-link) and negative (cannot-link) constraints find linear transformations or equivalently global Mahalanobis metrics. Additionally, they find metrics only according to the data points appearing in constraints (without considering other data points). In this paper, we consider the topological structure of data along with both positive and negative constraints. We propose a kernel-based metric learning method that provides a non-linear transformation. Experimental results on synthetic and real-world data sets show the effectiveness of our metric learning method.

In this paper, we present a novel approach to the evaluation of propositional answer-set programs. In particular, for programs with bounded treewidth, our algorithm is capable of (i) computing the number of answer sets in linear time and (ii) enumerating all answer sets with linear delay. Our algorithm relies on dynamic programming, which so far has not been applied to ASP-problems. Therefore, our approach significantly differs from standard ASP-systems which implement techniques stemming from SAT or CSP, and thus usually do not exploit fixed parameter properties of the programs. We provide first experimental results which underline that, for programs with low treewidth, already a prototypical implementation is competitive compared to state-of-the-art systems.

Kernel discriminant analysis (KDA) is an effective approach for supervised nonlinear dimensionality reduction. Probabilistic models can be used with KDA to improve its robustness. However, the state of the art of such models could only handle binary class problems, which confines their application in many real world problems. To overcome this limitation, we propose a novel nonparametric probabilistic model based on Gaussian Process for KDA to handle multiclass problems. The model provides a novel Bayesian interpretation for KDA, which allows its parameters to be automatically tuned through the optimization of the marginal loglikelihood of the data. Empirical study demonstrates the efficacy of the proposed model.

Equilibrium or near-equilibrium solutions to very large extensive form games are often computed by using abstractions to reduce the game size. A common abstraction technique for games with a large number of available actions is to restrict the number of legal actions in every state. This method has been used to discover equilibrium solutions for the game of no-limit heads-up Texas Hold'em. When using a solution to an abstracted game to play one side in the un-abstracted (real) game, the real opponent actions may not correspond to actions in the abstracted game. The most popular method for handling this situation is to translate opponent actions in the real game to the closest legal actions in the abstracted game. We show that this approach can result in a very exploitable player and propose an alternative solution. We use probabilistic mapping to translate a real action into a probability distribution over actions, whose weights are determined by a similarity metric. We show that this approach significantly reduces the exploitability when using an abstract solution in the real game.

We investigate prediction of users' desktop activities in the Unix domain. The learning techniques we explore do not require explicit user teaching. We show that simple efficient many-class learning can perform well for action prediction, significantly improving over previously published results and baselines. This finding is promising for various human-computer interaction scenarios where a rich set of potentially predictive features is available, where there can be many different actions to predict, and where there can be considerable nonstationarity.

We propose an approach for automatically ranking structured documents applied to patent prior art search. Our model, SVM Patent Ranking (SVM_PR) incorporates margin constraints that directly capture the specificities of patent citation ranking. Our approach combines patent domain knowledge features with meta-score features from several different general Information Retrieval methods. The training algorithm is an extension of the Pegasos algorithm with performance guarantees, effectively handling hundreds of thousands of patent-pair judgements in a high dimensional feature space. Experiments on a homogeneous essential wireless patent dataset show that SVM_PR performs on average 30%-40% better than many other state-of-the-art general-purpose Information Retrieval methods in terms of the NDCG measure at different cut-off positions.