The Approximate Linear Programming (ALP) approach to value function approximation for MDPs is a parametric value function approximation method, in that it represents the value function as a linear combination of features which are chosen a priori. Choosing these features can be a difficult challenge in itself. One recent effort, Regularized Approximate Linear Programming (RALP), uses L1 regularization to address this issue by combining a large initial set of features with a regularization penalty that favors a smooth value function with few non-zero weights. Rather than using smoothness as a backhanded way of addressing the feature selection problem, this paper starts with smoothness and develops a non-parametric approach to ALP that is consistent with the smoothness assumption. We show that this new approach has some favorable practical and analytical properties in comparison to (R)ALP.

Applications that adapt to a particular end user often make inaccurate predictions during the early stages when training data is limited. Although an end user can improve the learning algorithm by labeling more training data, this process is time consuming and too ad hoc to target a particular area of inaccuracy. To solve this problem, we propose a new learning algorithm based on Locally Weighted Logistic Regression for feature labeling by end users, enabling them to point out which features are important for a class, rather than provide new training instances. In our user study, the first allowing ordinary end users to freely choose features to label directly from text documents, our algorithm was more effective than others at leveraging end users’ feature labels to improve the learning algorithm. Our results strongly suggest that allowing users to freely choose features to label is a promising method for allowing end users to improve learning algorithms effectively.

We describe the Fourier basis, a linear value function approximation scheme based on the Fourier series. We empirically demonstrate that it performs well compared to radial basis functions and the polynomial basis, the two most popular fixed bases for linear value function approximation, and is competitive with learned proto-value functions.

Multi-task learning has received increasing attention in the past decade. Many supervised multi-task learning methods have been proposed, while unsupervised multi-task learning is still a rarely studied problem. In this paper, we propose to learn a kernel for multi-task clustering. Our goal is to learn a Reproducing Kernel Hilbert Space, in which the geometric structure of the data in each task is preserved, while the data distributions of any two tasks are as close as possible. This is formulated as a unified kernel learning framework, under which we study two types of kernel learning: nonparametric kernel learning and spectral kernel design. Both types of kernel learning can be solved by linear programming. Experiments on several cross-domain text data sets demonstrate that kernel k-means on the learned kernel can achieve better clustering results than traditional single-task clustering methods. It also outperforms the newly proposed multi-task clustering method.

The problem of influence maximization, i.e., mining top-k influential nodes from a social network such that the spread of influence in the network is maximized, is NP-hard. Most of the existing algorithms for the prob- lem are based on greedy algorithm. Although greedy algorithm can achieve a good approximation, it is computational expensive. In this paper, we propose a totally different approach based on Simulated Annealing(SA) for the influence maximization problem. This is the first SA based algorithm for the problem. Additionally, we propose two heuristic methods to accelerate the con- vergence process of SA, and a new method of comput- ing influence to speed up the proposed algorithm. Experimental results on four real networks show that the proposed algorithms run faster than the state-of-the-art greedy algorithm by 2-3 orders of magnitude while being able to improve the accuracy of greedy algorithm.

In many real world applications, the number of examples to learn from is plentiful, but we can only obtain limited information on each individual example. We study the possibilities of efficient, provably correct, large-scale learning in such settings. The main theme we would like to establish is that large amounts of examples can compensate for the lack of full information on each individual example. The type of partial information we consider can be due to inherent noise or from constraints on the type of interaction with the data source. In particular, we describe and analyze algorithms for budgeted learning, in which the learner can only view a few attributes of each training example, and algorithms for learning kernel-based predictors, when individual examples are corrupted by random noise.

With smart-phones becoming increasingly commonplace, there has been a subsequent surge in applications that continuously track the location of users. However, serious privacy concerns arise as people start to widely adopt these applications. Users will need to maintain policies to determine under which circumstances to share their location. Specifying these policies however, is a cumbersome task, suggesting that machine learning might be helpful. In this paper, we present a user-controllable method for learning location sharing policies. We use a classifier based on multivariate Gaussian mixtures that is suitably modified so as to restrict the evolution of the underlying policy to favor incremental and therefore human-understandable changes as new data arrives. We evaluate the model on real location-sharing policies collected from a live location-sharing social network, and we show that our method can learn policies in a user-controllable setting that are just as accurate as policies that do not evolve incrementally. Additionally, we highlight the strength of the generative modeling approach we take, by showing how our model easily extends to the semi-supervised setting.

In order to better protect and conserve biodiversity, ecologists use machine learning and statistics to understand how species respond to their environment and to predict how they will respond to future climate change, habitat loss and other threats. A fundamental modeling task is to estimate the probability that a given species is present in (or uses) a site, conditional on environmental variables such as precipitation and temperature. For a limited number of species, survey data consisting of both presence and absence records are available, and can be used to fit a variety of conventional classification and regression models. For most species, however, the available data consist only of occurrence records --- locations where the species has been observed. In two closely-related but separate bodies of ecological literature, diverse special-purpose models have been developed that contrast occurrence data with a random sample of available environmental conditions. The most widespread statistical approaches involve either fitting an exponential model of species' conditional probability of presence, or fitting a naive logistic model in which the random sample of available conditions is treated as absence data; both approaches have well-known drawbacks, and do not necessarily produce valid probabilities. After summarizing existing methods, we overcome their drawbacks by introducing a new scaled binomial loss function for estimating an underlying logistic model of species presence/absence. Like the Expectation-Maximization approach of Ward et al. and the method of Steinberg and Cardell, our approach requires an estimate of population prevalence, $\Pr(y=1)$, since prevalence is not identifiable from occurrence data alone. In contrast to the latter two methods, our loss function is straightforward to integrate into a variety of existing modeling frameworks such as generalized linear and additive models and boosted regression trees. We also demonstrate that approaches by Lele and Keim and by Lancaster and Imbens that surmount the identifiability issue by making parametric data assumptions do not typically produce valid probability estimates.

Query recommendation can not only effectively facilitate users to obtain their desired information but alsoincrease ads’ click-through rates. This paper presentsa general and highly efficient method for query recommendation. Given query sessions, we automatically generate many similar and dissimilar query-pairs as the prior knowledge. Then we learn a transformation from the prior knowledge to move similar queries closer such that similar queries tend to have similar hash values.This is formulated as minimizing the empirical error on the prior knowledge while maximizing the gap between the data and some partition hyperplanes randomly generated in advance. In the recommendation stage, we search queries that have similar hash values to the given query, rank the found queries and return the top K queries as the recommendation result. All the experimental results demonstrate that our method achieves encouraging results in terms of efficiency and recommendation performance.

K-flats is a model-based linear manifold clustering algorithm which has been successfully applied in many real-world scenarios. Though some previous works have shown that K-flats doesn’t always provide good performance, little effort has been devoted to analyze its inherent deficiency. In this paper, we address this challenge by showing that the deteriorative performance of K-flats can be attributed to the usual reconstruction error measure and the infinitely extending representations of linear models. Then we propose Localized K-flats algorithm (LKF), which introduces localized representations of linear models and a new distortion measure, to remove confusion among different clusters. Experiments on both synthetic and real-world data sets demonstrate the efficiency of the proposed algorithm. Moreover, preliminary experiments show that LKF has the potential to group manifolds with nonlinear structure.