Interactive dynamic influence diagrams (I-DIDs) are graphical models for  sequential  decision making  in  partially observable  settings shared  by other  agents.  Algorithms  for solving  I-DIDs  face the challenge  of an  exponentially  growing space  of candidate  models ascribed to  other agents, over time. Previous  approach for exactly   solving I-DIDs groups together  models having similar solutions into behaviorally  equivalent  classes  and  updates these  classes.   We present a  new method that, in addition  to aggregating behaviorally equivalent  models, further groups  models that  prescribe identical actions at a single time step. We show how to update these augmented classes  and prove  that  our  method is  exact.   The new  approach   enables us to bound the aggregated model space by the cardinality of   other  agents'  actions. We  evaluate  its  performance and  provide   empirical results in support.

Web-scale data has been used in a diverse range of language research. Most of this research has used web counts for only short, fixed spans of context. We present a unified view of using web counts for lexical disambiguation. Unlike previous approaches, our supervised and unsupervised systems combine information from multiple and overlapping segments of context. On the tasks of preposition selection and context-sensitive spelling correction, the supervised system reduces disambiguation error by 20-24% over the current state-of-the-art.

We propose a probabilistic transfer learning model that uses task-level features to control the task mixture selection in a hierarchical Bayesian model. These task-level features, although rarely used in existing approaches, can provide additional information to model complex task distributions and allow effective transfer to new tasks especially when only limited number of data are available. To estimate the model parameters, we develop an empirical Bayes method based on variational approximation techniques. Our experiments on information retrieval show that the proposed model achieves significantly better performance compared with other transfer learning methods.

Recent results have shown the interest  of trees-of-BDDs as a suitable target language for propositional knowledge compilation from the practical side. In the present paper,  the concept of tree-of-BDDs is extended to additional classes of data structures C, thus leading to trees-of-C representations (ToC). We provide a number of generic results enabling one to determine the queries/transformations satisfied by ToC depending on those satisfied by C. We also present some results about the spatial efficiency of the ToC languages. Focusing on the ToB language (and other related languages), we address a number of issues that remained open in (Subbarayan et al 2007). We show that beyond co and va, the ToB fragment satisfies im and me but satisfies neither cd nor any query among ce, se unlesssf P = NP. Among other results, we prove that ToB is not comparable w.r.t. succinctness with any of cnf, dnf,  Dnnf unless the polynomial hierarchy collapses.This contributes to the explanation of  some empirical results reported in (Subbarayan et al 2007).

Graphical games provide compact representation of a multiagent interaction when agents' payoffs depend only on actions of agents in their local neighborhood. We formally describe the problem of learning a graphical game model from limited observation of the payoff function, define three performance metrics for evaluating learned games, and investigate several learning algorithms based on minimizing empirical loss. Our first algorithm is a branch-and-bound search, which takes advantage of the structure of the empirical loss function to derive upper and lower bounds on loss at every node of the search tree. We also examine a greedy heuristic and local search algorithms. Our experiments with directed graphical games show that (i) when only a small sample of profile payoffs is available, branch-and-bound significantly outperforms other methods, and has competitive running time, but (ii) when many profiles are observed, greedy is nearly optimal and considerably better than other methods, at a fraction of branch-and-bound's running time. The results are comparable for undirected graphical games and when payoffs are sampled with noise.

We consider the problem of tactical assault planning in real-time strategy games where a team of friendly agents must launch an assault on an enemy. This problem offers many challenges including a highly dynamic and uncertain environment, multiple agents, durative actions, numeric attributes, and different optimization objectives. While the dynamics of this problem are quite complex, it is often possible to provide or learn a coarse simulation-based model of a tactical domain, which makes Monte-Carlo planning an attractive approach. In this paper, we investigate the use of UCT, a recent Monte-Carlo planning algorithm for this problem. UCT has recently shown impressive successes in the area of games, particularly Go, but has not yet been considered in the context of multi-agent tactical planning. We discuss the challenges of adapting UCT to our domain and an implementation which allows for the optimization of user specified objective functions. We present an evaluation of our approach on a range of tactical assault problems with different objectives in the RTS game Wargus. The results indicate that our planner is able to generate superior plans compared to several baselines and a human player.

Selecting important features in non-linear or kernel spaces is a difficult challenge in both classification and regression problems. When many of the features are irrelevant, kernel methods such as the support vector machine and kernel ridge regression can sometimes perform poorly. We propose weighting the features within a kernel with a sparse set of weights that are estimated in conjunction with the original classification or regression problem. The iterative algorithm, KNIFE, alternates between finding the coefficients of the original problem and finding the feature weights through kernel linearization. In addition, a slight modification of KNIFE yields an efficient algorithm for finding feature regularization paths, or the paths of each feature's weight. Simulation results demonstrate the utility of KNIFE for both kernel regression and support vector machines with a variety of kernels. Feature path realizations also reveal important non-linear correlations among features that prove useful in determining a subset of significant variables. Results on vowel recognition data, Parkinson's disease data, and microarray data are also given.

In many real world planning domains, some observation information is optional and useless to the execution of a plan; on the other hand, information acquisition may require some kind of cost. The problem of observation reduction for strong plans has been addressed in the literature. However, observation reduction for plans with contexts (which are more general and useful than strong plans in robotics) is still a open problem. In this paper, we present an attempt to solve the problem. Our first contribution is the definition of structured plans, which can encode sequential, conditional and iterative behaviors, and is expressive enough for dealing with incomplete observation information and internal states of the agent. A second contribution is an observation reduction algorithm for plans with contexts, which can transform a plan with contexts into a structured plan that only branches on necessary observation information.

Gaussian Processes (GPs) are promising Bayesian methods for classification and regression problems. They have also been used for semi-supervised learning tasks. In this paper, we propose a new algorithm for solving semi-supervised binary classification problem using sparse GP regression (GPR) models. It is closely related to semi-supervised learning based on support vector regression (SVR) and maximum margin clustering. The proposed algorithm is simple and easy to implement. It gives a sparse solution directly unlike the SVR based algorithm. Also, the hyperparameters are estimated easily without resorting to expensive cross-validation technique. Use of sparse GPR model helps in making the proposed algorithm scalable. Preliminary results on synthetic and real-world data sets demonstrate the efficacy of the new algorithm.

Gaussian Processes (GPs) are promising Bayesian methods for classification and regression problems. They have also been used for semi-supervised learning tasks. In this paper, we propose a new algorithm for solving semi-supervised binary classification problem using sparse GP regression (GPR) models. It is closely related to semi-supervised learning based on support vector regression (SVR) and maximum margin clustering. The proposed algorithm is simple and easy to implement. It gives a sparse solution directly unlike the SVR based algorithm. Also, the hyperparameters are estimated easily without resorting to expensive cross-validation technique. Use of sparse GPR model helps in making the proposed algorithm scalable. Preliminary results on synthetic and real-world data sets demonstrate the efficacy of the new algorithm.