Most previous heterogeneous transfer learning methods learn a cross-domain feature mapping between heterogeneous feature spaces based on a few cross-domain instance-correspondences, and these corresponding instances are assumed to be representative in the source and target domains respectively. However, in many real-world scenarios, this assumption may not hold. As a result, the constructed feature mapping may not be precisely due to the bias issue of the correspondences in the target or (and) source domain(s). In this case, a classifier trained on the labeled transformed-source-domain data may not be useful for the target domain. In this paper, we present a new transfer learning framework called Hybrid Heterogeneous Transfer Learning (HHTL), which allows the corresponding instances across domains to be biased in either the source or target domain. Specifically, we propose a deep learning approach to learn a feature mapping between cross-domain heterogeneous features as well as a better feature representation for mapped data to reduce the bias issue caused by the cross-domain correspondences. Extensive experiments on several multilingual sentiment classification tasks verify the effectiveness of our proposed approach compared with some baseline methods.

Recent work has shown that diverse teams can outperform a uniform team made of copies of the best agent. However, there are fundamental questions that were not asked before. When should we use diverse or uniform teams? How does the performance change as the action space or the teams get larger? Hence, we present a new model of diversity for teams, that is more general than previous models. We prove that the performance of a diverse team improves as the size of the action space gets larger. Concerning the size of the diverse team, we show that the performance converges exponentially fast to the optimal one as we increase the number of agents. We present synthetic experiments that allow us to gain further insights: even though a diverse team outperforms a uniform team when the size of the action space increases, the uniform team will eventually again play better than the diverse team for a large enough action space. We verify our predictions in a system of Go playing agents, where we show a diverse team that improves in performance as the board size increases, and eventually overcomes a uniform team.

This paper develops an efficient online algorithm for learning multiple consecutive tasks based on the K-SVD algorithm for sparse dictionary optimization. We first derive a batch multi-task learning method that builds upon K-SVD, and then extend the batch algorithm to train models online in a lifelong learning setting. The resulting method has lower computational complexity than other current lifelong learning algorithms while maintaining nearly identical model performance. Additionally, the proposed method offers an alternate formulation for lifelong learning that supports both task and feature similarity matrices.

Manifold learning is a powerful tool for solving nonlinear dimension reduction problems. By assuming that the high-dimensional data usually lie on a low-dimensional manifold, many algorithms have been proposed. However, most algorithms simply adopt the traditional graph Laplacian to encode the data locality, so the discriminative ability is limited and the embedding results are not always suitable for the subsequent classification. Instead, this paper deploys the signed graph Laplacian and proposes Signed Laplacian Embedding (SLE) for supervised dimension reduction. By exploring the label information, SLE comprehensively transfers the discrimination carried by the original data to the embedded low-dimensional space. Without perturbing the discrimination structure, SLE also retains the locality.Theoretically, we prove the immersion property by computing the rank of projection, and relate SLE to existing algorithms in the frame of patch alignment. Thorough empirical studies on synthetic and real datasets demonstrate the effectiveness of SLE.

Linear subspace is an important representation for many kinds of real-world data in computer vision and pattern recognition, e.g. faces, motion videos, speeches. In this paper, first we define pairwise angular similarity and angular distance for linear subspaces. The angular distance satisfies non-negativity, identity of indiscernibles, symmetry and triangle inequality, and thus it is a metric. Then we propose a method to compress linear subspaces into compact similarity-preserving binary signatures, between which the normalized Hamming distance is an unbiased estimator of the angular distance. We provide a lower bound on the length of the binary signatures which suffices to guarantee uniform distance-preservation within a set of subspaces. Experiments on face recognition demonstrate the effectiveness of the binary signature in terms of recognition accuracy, speed and storage requirement. The results show that, compared with the exact method, the approximation with the binary signatures achieves an order of magnitude speed-up, while requiring significantly smaller amount of storage space, yet it still accurately preserves the similarity, and achieves high recognition accuracy comparable to the exact method in face recognition.

A common bottleneck in deploying supervised learning systems is collecting human-annotated examples. In many domains, annotators form an opinion about the label of an example incrementally -- e.g., each additional word read from a document or each additional minute spent inspecting a video helps inform the annotation. In this paper, we investigate whether we can train learning systems more efficiently by requesting an annotation before inspection is fully complete -- e.g., after reading only 25 words of a document. While doing so may reduce the overall annotation time, it also introduces the risk that the annotator might not be able to provide a label if interrupted too early. We propose an anytime active learning approach that optimizes the annotation time and response rate simultaneously. We conduct user studies on two document classification datasets and develop simulated annotators that mimic the users. Our simulated experiments show that anytime active learning outperforms several baselines on these two datasets. For example, with an annotation budget of one hour, training a classifier by annotating the first 25 words of each document reduces classification error by 17% over annotating the first 100 words of each document.

Monte Carlo tree search (MCTS) algorithms are a popular approach to online decision-making in Markov decision processes (MDPs). These algorithms can, however, perform poorly in MDPs with high stochastic branching factors. In this paper, we study state aggregation as a way of reducing stochastic branching in tree search. Prior work has studied formal properties of MDP state aggregation in the context of dynamic programming and reinforcement learning, but little attention has been paid to state aggregation in MCTS. Our main result is a performance loss bound for a class of value function-based state aggregation criteria in expectimax search trees. We also consider how to construct MCTS algorithms that operate in the abstract state space but require a simulator of the ground dynamics only. We find that trajectory sampling algorithms like UCT can be adapted easily, but that sparse sampling algorithms present difficulties. As a proof of concept, we experimentally confirm that state aggregation can improve the finite-sample performance of UCT.

This paper studies the active learning problem in crowdsourcing settings, where multiple imperfect annotators with varying levels of expertise are available for labeling the data in a given task. Annotations collected from these labelers may be noisy and unreliable, and the quality of labeled data needs to be maintained for data mining tasks. Previous solutions have attempted to estimate individual users' reliability based on existing knowledge in each task, but for this to be effective each task requires a large quantity of labeled data to provide accurate estimates. In practice, annotation budgets for a given task are limited, so each instance can be presented to only a few users, each of whom can only label a few examples. To overcome data scarcity we propose a new probabilistic model that transfers knowledge from abundant unlabeled data in auxiliary domains to help estimate labelers' expertise. Based on this model we present a novel active learning algorithm that: a) simultaneously selects the most informative example and b) queries its label from the labeler with the best expertise. Experiments on both text and image datasets demonstrate that our proposed method outperforms other state-of-the-art active learning methods.

Scheduling under uncertainty is essential to many autonomous systems and logistics tasks. Probabilistic methods for solving temporal problems exist which quantify and attempt to minimize the probability of schedule failure. These methods are overly conservative, resulting in a loss in schedule utility. Chance constrained formalism address over-conservatism by imposing bounds on risk, while maximizing utility subject to these risk bounds. In this paper we present the probabilistic Simple Temporal Network (pSTN), a probabilistic formalism for representing temporal problems with bounded risk and a utility over event timing. We introduce a constrained optimisation algorithm for pSTNs that achieves compactness and efficiency through a problem encoding in terms of a parameterised STNU and its reformulation as a parameterised STN. We demonstrate through a car sharing application that our chance-constrained approach runs in the same time as the previous probabilistic approach, yields solutions with utility improvements of at least 5% over previous arts, while guaranteeing operation within the specified risk bound.

Hashing is a popular approximate nearest neighbor search approach for large-scale image retrieval. Supervised hashing, which incorporates similarity/dissimilarity information on entity pairs to improve the quality of hashing function learning, has recently received increasing attention. However, in the existing supervised hashing methods for images, an input image is usually encoded by a vector of hand-crafted visual features. Such hand-crafted feature vectors do not necessarily preserve the accurate semantic similarities of images pairs, which may often degrade the performance of hashing function learning. In this paper, we propose a supervised hashing method for image retrieval, in which we automatically learn a good image representation tailored to hashing as well as a set of hash functions. The proposed method has two stages. In the first stage, given the pairwise similarity matrix $S$ over training images, we propose a scalable coordinate descent method to decompose $S$ into a product of $HH^T$ where $H$ is a matrix with each of its rows being the approximate hash code associated to a training image. In the second stage, we propose to simultaneously learn a good feature representation for the input images as well as a set of hash functions, via a deep convolutional network tailored to the learned hash codes in $H$ and optionally the discrete class labels of the images. Extensive empirical evaluations on three benchmark datasets with different kinds of images show that the proposed method has superior performance gains over several state-of-the-art supervised and unsupervised hashing methods.