Recent several years have witnessed the surge of asynchronous (async-) parallel computing methods due to the extremely big data involved in many modern applications and also the advancement of multi-core machines and computer clusters. In optimization, most works about async-parallel methods are on unconstrained problems or those with block separable constraints. In this paper, we propose an async-parallel method based on block coordinate update (BCU) for solving convex problems with nonseparable linear constraint. Running on a single node, the method becomes a novel randomized primal-dual BCU with adaptive stepsize for multi-block affinely constrained problems. For these problems, Gauss-Seidel cyclic primal-dual BCU needs strong convexity to have convergence. On the contrary, merely assuming convexity, we show that the objective value sequence generated by the proposed algorithm converges in probability to the optimal value and also the constraint residual to zero. In addition, we establish an ergodic $O(1/k)$ convergence result, where $k$ is the number of iterations. Numerical experiments are performed to demonstrate the efficiency of the proposed method and significantly better speed-up performance than its sync-parallel counterpart.

We consider the problem of efficient randomized dimensionality reduction with norm-preservation guarantees. Specifically we prove data-dependent Johnson-Lindenstrauss-type geometry preservation guarantees for Ho's random subspace method: When data satisfy a mild regularity condition -- the extent of which can be estimated by sampling from the data -- then random subspace approximately preserves the Euclidean geometry of the data with high probability. Our guarantees are of the same order as those for random projection, namely the required dimension for projection is logarithmic in the number of data points, but have a larger constant term in the bound which depends upon this regularity. A challenging situation is when the original data have a sparse representation, since this implies a very large projection dimension is required: We show how this situation can be improved for sparse binary data by applying an efficient `densifying' preprocessing, which neither changes the Euclidean geometry of the data nor requires an explicit matrix-matrix multiplication. We corroborate our theoretical findings with experiments on both dense and sparse high-dimensional datasets from several application domains.

Multi-armed bandits are a quintessential machine learning problem requiring the balancing of exploration and exploitation. While there has been progress in developing algorithms with strong theoretical guarantees, there has been less focus on practical near-optimal finite-time performance. In this paper, we propose an algorithm for Bayesian multi-armed bandits that utilizes value-function-driven online planning techniques. Building on previous work on UCB and Gittins index, we introduce linearly-separable value functions that take both the expected return and the benefit of exploration into consideration to perform n-step lookahead. The algorithm enjoys a sub-linear performance guarantee and we present simulation results that confirm its strength in problems with structured priors. The simplicity and generality of our approach makes it a strong candidate for analyzing more complex multi-armed bandit problems.

Recent approaches based on artificial neural networks (ANNs) have shown promising results for named-entity recognition (NER). In order to achieve high performances, ANNs need to be trained on a large labeled dataset. However, labels might be difficult to obtain for the dataset on which the user wants to perform NER: label scarcity is particularly pronounced for patient note de-identification, which is an instance of NER. In this work, we analyze to what extent transfer learning may address this issue. In particular, we demonstrate that transferring an ANN model trained on a large labeled dataset to another dataset with a limited number of labels improves upon the state-of-the-art results on two different datasets for patient note de-identification.

Conventional text classification models make a bag-of-words assumption reducing text, fundamentally a sequence of words, into word occurrence counts per document. Recent algorithms such as word2vec and fastText are capable of learning semantic meaning and similarity between words in an entirely unsupervised manner using a contextual window and doing so much faster than previous methods. Each word is represented as a vector such that similar meaning words such as 'strong' and 'powerful' are in the same general Euclidian space. Open questions about these embeddings include their usefulness across classification tasks and the optimal set of documents to build the embeddings. In this work, we demonstrate the usefulness of embeddings for improving the state of the art in classification for our tasks and demonstrate that specific word embeddings built in the domain and for the tasks can improve performance over general word embeddings (learnt on news articles, Wikipedia or PubMed).

Linking human whole-body motion and natural language is of great interest for the generation of semantic representations of observed human behaviors as well as for the generation of robot behaviors based on natural language input. While there has been a large body of research in this area, most approaches that exist today require a symbolic representation of motions (e.g. in the form of motion primitives), which have to be defined a-priori or require complex segmentation algorithms. In contrast, recent advances in the field of neural networks and especially deep learning have demonstrated that sub-symbolic representations that can be learned end-to-end usually outperform more traditional approaches, for applications such as machine translation. In this paper we propose a generative model that learns a bidirectional mapping between human whole-body motion and natural language using deep recurrent neural networks (RNNs) and sequence-to-sequence learning. Our approach does not require any segmentation or manual feature engineering and learns a distributed representation, which is shared for all motions and descriptions. We evaluate our approach on 2,846 human whole-body motions and 6,187 natural language descriptions thereof from the KIT Motion-Language Dataset. Our results clearly demonstrate the effectiveness of the proposed model: We show that our model generates a wide variety of realistic motions only from descriptions thereof in form of a single sentence. Conversely, our model is also capable of generating correct and detailed natural language descriptions from human motions.

Fingerprinting based WLAN indoor positioning system (FWIPS) provides a promising indoor positioning solution to meet the growing interests for indoor location-based services (e.g., indoor way finding or geo-fencing). FWIPS is preferred because it requires no additional infrastructure for deploying an FWIPS and achieving the position estimation by reusing the available WLAN and mobile devices, and capable of providing absolute position estimation. For fingerprinting based positioning (FbP), a model is created to provide reference values of observable features (e.g., signal strength from access point (AP)) as a function of location during offline stage. One widely applied method to build a complete and an accurate reference database (i.e. radio map (RM)) for FWIPS is carrying out a site survey throughout the region of interest (RoI). Along the site survey, the readings of received signal strength (RSS) from all visible APs at each reference point (RP) are collected. This site survey, however, is time-consuming and labor-intensive, especially in the case that the RoI is large (e.g., an airport or a big mall). This bottleneck hinders the wide commercial applications of FWIPS (e.g., proximity promotions in a shopping center). To diminish the cost of site survey, we propose a probabilistic model, which combines fingerprinting based positioning (FbP) and RM generation based on stochastic variational Bayesian inference (SVBI). This SVBI based position and RSS estimation has three properties: i) being able to predict the distribution of the estimated position and RSS, ii) treating each observation of RSS at each RP as an example to learn for FbP and RM generation instead of using the whole RM as an example, and iii) requiring only one time training of the SVBI model for both localization and RSS estimation. These benefits make it outperforms the previous proposed approaches.

Principal Component Analysis (PCA) is a very successful dimensionality reduction technique, widely used in predictive modeling. A key factor in its widespread use in this domain is the fact that the projection of a dataset onto its first $K$ principal components minimizes the sum of squared errors between the original data and the projected data over all possible rank $K$ projections. Thus, PCA provides optimal low-rank representations of data for least-squares linear regression under standard modeling assumptions. On the other hand, when the loss function for a prediction problem is not the least-squares error, PCA is typically a heuristic choice of dimensionality reduction -- in particular for classification problems under the zero-one loss. In this paper we target classification problems by proposing a straightforward alternative to PCA that aims to minimize the difference in margin distribution between the original and the projected data. Extensive experiments show that our simple approach typically outperforms PCA on any particular dataset, in terms of classification error, though this difference is not always statistically significant, and despite being a filter method is frequently competitive with Partial Least Squares (PLS) and Lasso on a wide range of datasets.

In a series of recent works, we have generalised the consistency results in the stochastic block model literature to the case of uniform and non-uniform hypergraphs. The present paper continues the same line of study, where we focus on partitioning weighted uniform hypergraphs---a problem often encountered in computer vision. This work is motivated by two issues that arise when a hypergraph partitioning approach is used to tackle computer vision problems: (i) The uniform hypergraphs constructed for higher-order learning contain all edges, but most have negligible weights. Thus, the adjacency tensor is nearly sparse, and yet, not binary. (ii) A more serious concern is that standard partitioning algorithms need to compute all edge weights, which is computationally expensive for hypergraphs. This is usually resolved in practice by merging the clustering algorithm with a tensor sampling strategy---an approach that is yet to be analysed rigorously. We build on our earlier work on partitioning dense unweighted uniform hypergraphs (Ghoshdastidar and Dukkipati, ICML, 2015), and address the aforementioned issues by proposing provable and efficient partitioning algorithms. Our analysis justifies the empirical success of practical sampling techniques. We also complement our theoretical findings by elaborate empirical comparison of various hypergraph partitioning schemes.

We develop a new method called Discriminated Hub Graphical Lasso (DHGL) based on Hub Graphical Lasso (HGL) by providing prior information of hubs. We apply this new method in two situations: with known hubs and without known hubs. Then we compare DHGL with HGL using several measures of performance. When some hubs are known, we can always estimate the precision matrix better via DHGL than HGL. When no hubs are known, we use Graphical Lasso (GL) to provide information of hubs and find that the performance of DHGL will always be better than HGL if correct prior information is given and will seldom degenerate when the prior information is wrong.