We aim to reduce the burden of programming and deploying autonomous systems to work in concert with people in time-critical domains, such as military field operations and disaster response. Deployment plans for these operations are frequently negotiated on-the-fly by teams of human planners. A human operator then translates the agreed upon plan into machine instructions for the robots. We present an algorithm that reduces this translation burden by inferring the final plan from a processed form of the human team's planning conversation. Our approach combines probabilistic generative modeling with logical plan validation used to compute a highly structured prior over possible plans. This hybrid approach enables us to overcome the challenge of performing inference over the large solution space with only a small amount of noisy data from the team planning session. We validate the algorithm through human subject experimentation and show we are able to infer a human team's final plan with 83% accuracy on average. We also describe a robot demonstration in which two people plan and execute a first-response collaborative task with a PR2 robot. To the best of our knowledge, this is the first work that integrates a logical planning technique within a generative model to perform plan inference.

Recommender systems, especially the newly launched ones, have to deal with the data-sparsity issue, where little existing rating information is available. Recently, transfer learning has been proposed to address this problem by leveraging the knowledge from related recommender systems where rich collaborative data are available. However, most previous transfer learning models assume that entity-correspondences across different systems are given as input, which means that for any entity (e.g., a user or an item) in a target system, its corresponding entity in a source system is known. This assumption can hardly be satisfied in real-world scenarios where entity-correspondences across systems are usually unknown, and the cost of identifying them can be expensive. For example, it is extremely difficult to identify whether a user A from Facebook and a user B from Twitter are the same person. In this paper, we propose a framework to construct entity correspondence with limited budget by using active learning to facilitate knowledge transfer across recommender systems. Specifically, for the purpose of maximizing knowledge transfer, we first iteratively select entities in the target system based on our proposed criterion to query their correspondences in the source system. We then plug the actively constructed entity-correspondence mapping into a general transferred collaborative-filtering model to improve recommendation quality. We perform extensive experiments on real world datasets to verify the effectiveness of our proposed framework for this cross-system recommendation problem.

In AI and Web communities, modularity-based graph clustering algorithms are being applied to various applications. However, existing algorithms are not applied to large graphs because they have to scan all vertices/edges iteratively. The goal of this paper is to efficiently compute clusters with high modularity from extremely large graphs with more than a few billion edges. The heart of our solution is to compute clusters by incrementally pruning unnecessary vertices/edges and optimizing the order of vertex selections. Our experiments show that our proposal outperforms all other modularity-based algorithms in terms of computation time, and it finds clusters with high modularity.

The minimax concave plus penalty (MCP) has been demonstrated to be effective in nonconvex penalization for feature selection. In this paper we propose a novel construction approach for MCP. In particular, we show that MCP can be derived from a concave conjugate of the Euclidean distance function. This construction approach in turn leads us to an augmented Lagrange multiplier method for solving the penalized regression problem with MCP. In our method each tuning parameter corresponds to a feature, and these tuning parameters can be automatically updated. We also develop a d.c. (difference of convex functions) programming approach for the penalized regression problem. We find that the augmented Lagrange multiplier method degenerates into the d.c. programming method under specific conditions. Experimental analysis is conducted on a set of simulated data. The result is encouraging.

A significant problem of Gaussian process (GP) is its unfavorable scaling with a large amount of data. To overcome this issue, we present a novel GP approximation scheme for online regression. Our model is based on a combination of multiple GPs with random hyperparameters. The model is trained by incrementally allocating new examples to a selected subset of GPs. The selection is carried out efficiently by optimizing a submodular function. Experiments on real-world data sets showed that our method outperforms existing online GP regression methods in both accuracy and efficiency. The applicability of the proposed method is demonstrated by the mouse-trajectory prediction in an Internet banking scenario.

Non-negative tensor factorization (NTF) has attracted great attention in the machine learning community. In this paper, we extend traditional non-negative tensor factorization into a supervised discriminative decomposition, referred as Supervised Non-negative Tensor Factorization with Maximum-Margin Constraint(SNTFM2). SNTFM2 formulates the optimal discriminative factorization of non-negative tensorial data as a coupled least-squares optimization problem via a maximum-margin method. As a result, SNTFM2 not only faithfully approximates the tensorial data by additive combinations of the basis, but also obtains a strong generalization power to discriminative analysis (in particularfor classification in this paper). The experimental results show the superiority of our proposed model over state-of-the-art techniques on both toy and real world data sets.

Although the heuristic search algorithm A* is well-known to be optimally efficient, this result explicitly assumes forward search. Bidirectional search has long held promise for surpassing A*'s efficiency, and many varieties have been proposed, but it has proven difficult to achieve robust performance across multiple domains in practice. We introduce a simple bidirectional search technique called Incremental KKAdd that judiciously performs backward search to improve the accuracy of the forward heuristic function for any search algorithm. We integrate this technique with A*, assess its theoretical properties, and empirically evaluate its performance across seven benchmark domains. In the best case, it yields a factor of six reduction in node expansions and CPU time compared to A*, and in the worst case, its overhead is provably bounded by a user-supplied parameter, such as 1%. Viewing performance across all domains, it also surpasses previously proposed bidirectional search algorithms. These results indicate that Incremental KKAdd is a robust way to leverage bidirectional search in practice.

While much research exists on resolving spatial natural language references to known locations, little work deals with handling references to unknown locations. In this paper we introduce and evaluate algorithms integrated into a cognitive architecture which allow an agent to learn about its environ-ment while resolving references to both known and unknown locations. We also describe how multiple components in the architecture jointly facilitate these capabilities.

Stochastic local search (SLS) is the dominant paradigm for incomplete SAT and MAXSAT solvers. Early studies on small 3SAT instances found that the use of “best improving” moves did not improve search compared to using an arbitrary “first improving” move. Yet SLS algorithms continue to use best improving moves. We revisit this issue by studying very large random and industrial MAXSAT problems. Because locating best improving moves is more expensive than first improving moves, we designed an “approximate best” improving move algorithm and prove that it is as efficient as first improving move SLS. For industrial problems the first local optima found using best improving moves are statistically significantly better than local optima found using first improving moves. However, this advantage reverses as search continues and algorithms must explore equal moves on plateaus. This reversal appears to be associated with critical variables that are in many clauses and that also yield large improving moves.

In many tasks related to reasoning about consequences of a logical theory, it is desirable to decompose the theory into a number of components with weakly-related or independent signatures. This facilitates reasoning when signature of a query formula belongs to only one of the components. However, an initial theory may be subject to change due to execution of actions affecting features mentioned in the theory. Having once computed a decomposition of a theory, one would like to know whether a decomposition has to be computed again for the theory obtained from taking into account the changes resulting from execution of an action. In the paper, we address this problem in the scope of the situation calculus, where change of an initial theory is related to the well-studied notion of progression. Progression provides a form of forward reasoning; it relies on forgetting values of those features which are subject to change and computing new values for them. We prove new results about properties of decomposition components under forgetting and show when a decomposition can be preserved in progression of an initial theory.