arXiv subjects

Hao Liu

Publications and source records attributed to Hao Liu.

At least 595 records · Page 33Linked to original sources

arXiv2020

Doubly Robust Off-Policy Learning on Low-Dimensional Manifolds by Deep Neural Networks

Causal inference explores the causation between actions and the consequent rewards on a covariate set. Recently deep learning has achieved a remarkable performance in causal inference, but existing statistical theories cannot well explain such an empirical success, especially when the covariates are high-dimensional. Most theoretical results in causal inference are asymptotic, suffer from the curse of dimensionality, and only work for the finite-action scenario. To bridge such a gap between theory and practice, this paper studies doubly robust off-policy learning by deep neural networks. When the covariates lie on a low-dimensional manifold, we prove nonasymptotic regret bounds, which converge at a fast rate depending on the intrinsic dimension of the manifold. Our results cover both the finite- and continuous-action scenarios. Our theory shows that deep neural networks are adaptive to the low-dimensional geometric structures of the covariates, and partially explains the success of deep learning for causal inference.

cs.LG↗

arXiv2020

Interactive Reinforcement Learning for Feature Selection with Decision Tree in the Loop

We study the problem of balancing effectiveness and efficiency in automated feature selection. After exploring many feature selection methods, we observe a computational dilemma: 1) traditional feature selection is mostly efficient, but difficult to identify the best subset; 2) the emerging reinforced feature selection automatically navigates to the best subset, but is usually inefficient. Can we bridge the gap between effectiveness and efficiency under automation? Motivated by this dilemma, we aim to develop a novel feature space navigation method. In our preliminary work, we leveraged interactive reinforcement learning to accelerate feature selection by external trainer-agent interaction. In this journal version, we propose a novel interactive and closed-loop architecture to simultaneously model interactive reinforcement learning (IRL) and decision tree feedback (DTF). Specifically, IRL is to create an interactive feature selection loop and DTF is to feed structured feature knowledge back to the loop. First, the tree-structured feature hierarchy from decision tree is leveraged to improve state representation. In particular, we represent the selected feature subset as an undirected graph of feature-feature correlations and a directed tree of decision features. We propose a new embedding method capable of empowering graph convolutional network to jointly learn state representation from both the graph and the tree. Second, the tree-structured feature hierarchy is exploited to develop a new reward scheme. In particular, we personalize reward assignment of agents based on decision tree feature importance. In addition, observing agents' actions can be feedback, we devise another reward scheme, to weigh and assign reward based on the feature selected frequency ratio in historical action records. Finally, we present extensive experiments on real-world datasets to show the improved performance.

cs.LG↗

arXiv2020

AutoFS: Automated Feature Selection via Diversity-aware Interactive Reinforcement Learning

In this paper, we study the problem of balancing effectiveness and efficiency in automated feature selection. Feature selection is a fundamental intelligence for machine learning and predictive analysis. After exploring many feature selection methods, we observe a computational dilemma: 1) traditional feature selection methods (e.g., mRMR) are mostly efficient, but difficult to identify the best subset; 2) the emerging reinforced feature selection methods automatically navigate feature space to explore the best subset, but are usually inefficient. Are automation and efficiency always apart from each other? Can we bridge the gap between effectiveness and efficiency under automation? Motivated by such a computational dilemma, this study is to develop a novel feature space navigation method. To that end, we propose an Interactive Reinforced Feature Selection (IRFS) framework that guides agents by not just self-exploration experience, but also diverse external skilled trainers to accelerate learning for feature exploration. Specifically, we formulate the feature selection problem into an interactive reinforcement learning framework. In this framework, we first model two trainers skilled at different searching strategies: (1) KBest based trainer; (2) Decision Tree based trainer. We then develop two strategies: (1) to identify assertive and hesitant agents to diversify agent training, and (2) to enable the two trainers to take the teaching role in different stages to fuse the experiences of the trainers and diversify teaching process. Such a hybrid teaching strategy can help agents to learn broader knowledge, and, thereafter, be more effective. Finally, we present extensive experiments on real-world datasets to demonstrate the improved performances of our method: more efficient than existing reinforced selection and more effective than classic selection.

cs.LG↗

arXiv2020

Curvature Regularized Surface Reconstruction from Point Cloud

We propose a variational functional and fast algorithms to reconstruct implicit surface from point cloud data with a curvature constraint. The minimizing functional balances the distance function from the point cloud and the mean curvature term. Only the point location is used, without any local normal or curvature estimation at each point. With the added curvature constraint, the computation becomes particularly challenging. To enhance the computational efficiency, we solve the problem by a novel operator splitting scheme. It replaces the original high-order PDEs by a decoupled PDE system, which is solved by a semi-implicit method. We also discuss approach using an augmented Lagrangian method. The proposed method shows robustness against noise, and recovers concave features and sharp corners better compared to models without curvature constraint. Numerical experiments in two and three dimensional data sets, noisy and sparse data are presented to validate the model.

cs.CV↗

arXiv2020

On the Numerical Solution of Nonlinear Eigenvalue Problems for the Monge-Ampère Operator

In this article, we report the results we obtained when investigating the numerical solution of some nonlinear eigenvalue problems for the Monge-Ampère operator $v\rightarrow \det \mathbf{D}^2 v$. The methodology we employ relies on the following ingredients: (i) A divergence formulation of the eigenvalue problems under consideration. (ii) The time discretization by operator-splitting of an initial value problem (a kind of gradient flow) associated with each eigenvalue problem. (iii) A finite element approximation relying on spaces of continuous piecewise affine functions. To validate the above methodology, we applied it to the solution of problems with known exact solutions: The results we obtained suggest convergence to the exact solution when the space discretization step $h\rightarrow 0$. We considered also test problems with no known exact solutions.

math.NA↗

arXiv2020

DARWIN: A Highly Flexible Platform for Imaging Research in Radiology

To conduct a radiomics or deep learning research experiment, the radiologists or physicians need to grasp the needed programming skills, which, however, could be frustrating and costly when they have limited coding experience. In this paper, we present DARWIN, a flexible research platform with a graphical user interface for medical imaging research. Our platform is consists of a radiomics module and a deep learning module. The radiomics module can extract more than 1000 dimension features(first-, second-, and higher-order) and provided many draggable supervised and unsupervised machine learning models. Our deep learning module integrates state of the art architectures of classification, detection, and segmentation tasks. It allows users to manually select hyperparameters, or choose an algorithm to automatically search for the best ones. DARWIN also offers the possibility for users to define a custom pipeline for their experiment. These flexibilities enable radiologists to carry out various experiments easily.

eess.IV↗

arXiv2020

Transmitting Extra Bits by Rotating Signal Constellations

In this letter, we propose a novel LDPC coding scheme to transmit extra bits aided by rotated signal constellations without any additional cost in transmission power or bandwidth. In the proposed scheme, the LDPC coded data are modulated by a rotated two-dimensional signal constellation, in which the rotation angle is specified by the given extra bits. At the receiver, the rotation angle is estimated with the aid of the statistical learning of the syndrome of the LDPC code. After recovering the rotation angle, the coded payload data can be decoded by the LDPC decoder. The simulation results show that, for an LDPC code of length 2304, up to four extra bits can be transmitted with negligible influence on the reliability of the LDPC coded data.

cs.IT↗

arXiv2020

Identifying and Tracking Solar Magnetic Flux Elements with Deep Learning

Deep learning has drawn a lot of interest in recent years due to its effectiveness in processing big and complex observational data gathered from diverse instruments. Here we propose a new deep learning method, called SolarUnet, to identify and track solar magnetic flux elements or features in observed vector magnetograms based on the Southwest Automatic Magnetic Identification Suite (SWAMIS). Our method consists of a data pre-processing component that prepares training data from the SWAMIS tool, a deep learning model implemented as a U-shaped convolutional neural network for fast and accurate image segmentation, and a post-processing component that prepares tracking results. SolarUnet is applied to data from the 1.6 meter Goode Solar Telescope at the Big Bear Solar Observatory. When compared to the widely used SWAMIS tool, SolarUnet is faster while agreeing mostly with SWAMIS on feature size and flux distributions, and complementing SWAMIS in tracking long-lifetime features. Thus, the proposed physics-guided deep learning-based tool can be considered as an alternative method for solar magnetic tracking.

astro-ph.SR↗

arXiv2020

Hybrid Discriminative-Generative Training via Contrastive Learning

Contrastive learning and supervised learning have both seen significant progress and success. However, thus far they have largely been treated as two separate objectives, brought together only by having a shared neural network. In this paper we show that through the perspective of hybrid discriminative-generative training of energy-based models we can make a direct connection between contrastive learning and supervised learning. Beyond presenting this unified view, we show our specific choice of approximation of the energy-based loss outperforms the existing practice in terms of classification accuracy of WideResNet on CIFAR-10 and CIFAR-100. It also leads to improved performance on robustness, out-of-distribution detection, and calibration.

cs.LG↗

arXiv2020

Polestar: An Intelligent, Efficient and National-Wide Public Transportation Routing Engine

Public transportation plays a critical role in people's daily life. It has been proven that public transportation is more environmentally sustainable, efficient, and economical than any other forms of travel. However, due to the increasing expansion of transportation networks and more complex travel situations, people are having difficulties in efficiently finding the most preferred route from one place to another through public transportation systems. To this end, in this paper, we present Polestar, a data-driven engine for intelligent and efficient public transportation routing. Specifically, we first propose a novel Public Transportation Graph (PTG) to model public transportation system in terms of various travel costs, such as time or distance. Then, we introduce a general route search algorithm coupled with an efficient station binding method for efficient route candidate generation. After that, we propose a two-pass route candidate ranking module to capture user preferences under dynamic travel situations. Finally, experiments on two real-world data sets demonstrate the advantages of Polestar in terms of both efficiency and effectiveness. Indeed, in early 2019, Polestar has been deployed on Baidu Maps, one of the world's largest map services. To date, Polestar is servicing over 330 cities, answers over a hundred millions of queries each day, and achieves substantial improvement of user click ratio.

cs.AI↗

arXiv2020

Deep Learning for 3D Point Clouds: A Survey

Point cloud learning has lately attracted increasing attention due to its wide applications in many areas, such as computer vision, autonomous driving, and robotics. As a dominating technique in AI, deep learning has been successfully used to solve various 2D vision problems. However, deep learning on point clouds is still in its infancy due to the unique challenges faced by the processing of point clouds with deep neural networks. Recently, deep learning on point clouds has become even thriving, with numerous methods being proposed to address different problems in this area. To stimulate future research, this paper presents a comprehensive review of recent progress in deep learning methods for point clouds. It covers three major tasks, including 3D shape classification, 3D object detection and tracking, and 3D point cloud segmentation. It also presents comparative results on several publicly available datasets, together with insightful observations and inspiring future research directions.

cs.CV↗

arXiv2020

SKEP: Sentiment Knowledge Enhanced Pre-training for Sentiment Analysis

Recently, sentiment analysis has seen remarkable advance with the help of pre-training approaches. However, sentiment knowledge, such as sentiment words and aspect-sentiment pairs, is ignored in the process of pre-training, despite the fact that they are widely used in traditional sentiment analysis approaches. In this paper, we introduce Sentiment Knowledge Enhanced Pre-training (SKEP) in order to learn a unified sentiment representation for multiple sentiment analysis tasks. With the help of automatically-mined knowledge, SKEP conducts sentiment masking and constructs three sentiment knowledge prediction objectives, so as to embed sentiment information at the word, polarity and aspect level into pre-trained sentiment representation. In particular, the prediction of aspect-sentiment pairs is converted into multi-label classification, aiming to capture the dependency between words in a pair. Experiments on three kinds of sentiment tasks show that SKEP significantly outperforms strong pre-training baseline, and achieves new state-of-the-art results on most of the test datasets. We release our code at https://github.com/baidu/Senta.

cs.CL↗

arXiv2020

Inferring Vector Magnetic Fields from Stokes Profiles of GST/NIRIS Using a Convolutional Neural Network

We propose a new machine learning approach to Stokes inversion based on a convolutional neural network (CNN) and the Milne-Eddington (ME) method. The Stokes measurements used in this study were taken by the Near InfraRed Imaging Spectropolarimeter (NIRIS) on the 1.6 m Goode Solar Telescope (GST) at the Big Bear Solar Observatory. By learning the latent patterns in the training data prepared by the physics-based ME tool, the proposed CNN method is able to infer vector magnetic fields from the Stokes profiles of GST/NIRIS. Experimental results show that our CNN method produces smoother and cleaner magnetic maps than the widely used ME method. Furthermore, the CNN method is 4~6 times faster than the ME method, and is able to produce vector magnetic fields in near real-time, which is essential to space weather forecasting. Specifically, it takes ~50 seconds for the CNN method to process an image of 720 x 720 pixels comprising Stokes profiles of GST/NIRIS. Finally, the CNN-inferred results are highly correlated to the ME-calculated results and are closer to the ME's results with the Pearson product-moment correlation coefficient (PPMCC) being closer to 1 on average than those from other machine learning algorithms such as multiple support vector regression and multilayer perceptrons (MLP). In particular, the CNN method outperforms the current best machine learning method (MLP) by 2.6% on average in PPMCC according to our experimental study. Thus, the proposed physics-assisted deep learning-based CNN tool can be considered as an alternative, efficient method for Stokes inversion for high resolution polarimetric observations obtained by GST/NIRIS.

astro-ph.SR↗

arXiv2020

Comparison of the Planck 2018 CMB polarization maps in the BICEP2/Keck region

We examine the statistical properties of polarization maps from Planck 2018 within the patch of sky observed by the BICEP2/Keck experiment using the one point distribution function (1PDF), skewness, and kurtosis statistics. Our analysis is performed for the Q and U Stokes parameters and for the corresponding E- and B-modes of the CMB signals. We extend our analysis by studying the correlations between CMB polarization maps and residual maps (the difference between the full signal and the CMB map) for the frequency range of 100--217 GHz with both the Q/U and E/B approaches. Although all the CMB maps reveal almost Gaussian statistical properties for Q/U and E/B domains, we have detected very significant anomalies for cross-correlations with residuals at 100 GHz at the level of $3.7σ$ for the Commander map and $5.2σ$ for NILC, for both the Q and U parameters. Using the NILC--Commander difference, which does not contain a cosmological signal, we find a sub-dominant non-Gaussian component in Q skewness and kurtosis at the level of $4.3σ$ and $10σ$, respectively. For the B-mode we have found a very high level of cross-correlation (0.63--0.69) between the NILC/Commander maps and the 143 GHz total signal, which cannot be associated with the cosmological component. These strong deviations suggest that remnants of foregrounds, systematic effects, and component separation exist in the 2018 Planck CMB polarization maps in the BICEP2 sky area, which is far away from the Galactic plane. Our analysis also demonstrates the preferability of the Q/U domain over E/B for determination of the statistical properties of the derived CMB signals, due to non-locality of the transition Q/U $\rightarrow$ E/B.

astro-ph.CO↗

arXiv2020

Nanometric precision distance metrology via chip-scale soliton microcombs

Laser interferometry serves a fundamental role in science and technology, assisting precision metrology and dimensional length measurement. During the past decade, laser frequency combs - a coherent optical-microwave frequency ruler over a broad spectral range with traceability to time-frequency standards - have contributed pivotal roles in laser dimensional metrology with ever-growing demands in measurement precision. Here we report spectrally-resolved laser dimensional metrology via a soliton frequency microcomb, with nanometric-scale precision. Spectral interferometry provides information on the optical time-of-flight signature, and the large free-spectral range and high-coherence of the microcomb enables tooth-resolved and high-visibility interferograms that can be directly readout with optical spectrum instrumentation. We employ a hybrid timing signal from comb-line homodyne interferometry and microcomb spectrally-resolved interferometry - all from the same spectral interferogram. Our combined soliton and homodyne architecture demonstrates a 3-nm repeatability achieved via homodyne interferometry, and over 1,000-seconds stability in the long-term precision metrology at the white noise limits.

physics.ins-det↗

arXiv2020

PuzzleNet: Scene Text Detection by Segment Context Graph Learning

Recently, a series of decomposition-based scene text detection methods has achieved impressive progress by decomposing challenging text regions into pieces and linking them in a bottom-up manner. However, most of them merely focus on linking independent text pieces while the context information is underestimated. In the puzzle game, the solver often put pieces together in a logical way according to the contextual information of each piece, in order to arrive at the correct solution. Inspired by it, we propose a novel decomposition-based method, termed Puzzle Networks (PuzzleNet), to address the challenging scene text detection task in this work. PuzzleNet consists of the Segment Proposal Network (SPN) that predicts the candidate text segments fitting arbitrary shape of text region, and the two-branch Multiple-Similarity Graph Convolutional Network (MSGCN) that models both appearance and geometry correlations between each segment to its contextual ones. By building segments as context graphs, MSGCN effectively employs segment context to predict combinations of segments. Final detections of polygon shape are produced by merging segments according to the predicted combinations. Evaluations on three benchmark datasets, ICDAR15, MSRA-TD500 and SCUT-CTW1500, have demonstrated that our method can achieve better or comparable performance than current state-of-the-arts, which is beneficial from the exploitation of segment context graph.

cs.CV↗

arXiv2020

Predicting Coronal Mass Ejections Using SDO/HMI Vector Magnetic Data Products and Recurrent Neural Networks

We present two recurrent neural networks (RNNs), one based on gated recurrent units and the other based on long short-term memory, for predicting whether an active region (AR) that produces an M- or X-class flare will also produce a coronal mass ejection (CME). We model data samples in an AR as time series and use the RNNs to capture temporal information of the data samples. Each data sample has 18 physical parameters, or features, derived from photospheric vector magnetic field data taken by the Helioseismic and Magnetic Imager (HMI) on board the Solar Dynamics Observatory (SDO). We survey M- and X-class flares that occurred from 2010 May to 2019 May using the Geostationary Operational Environmental Satellite's X-ray flare catalogs provided by the National Centers for Environmental Information (NCEI), and select those flares with identified ARs in the NCEI catalogs. In addition, we extract the associations of flares and CMEs from the Space Weather Database Of Notifications, Knowledge, Information (DONKI). We use the information gathered above to build the labels (positive versus negative) of the data samples at hand. Experimental results demonstrate the superiority of our RNNs over closely related machine learning methods in predicting the labels of the data samples. We also discuss an extension of our approach to predict a probabilistic estimate of how likely an M- or X-class flare will initiate a CME, with good performance results. To our knowledge this is the first time that RNNs have been used for CME prediction.

astro-ph.SR↗

arXiv2020

Statistics of CMB polarization angles

We study the distribution functions of the CMB polarization angle $ψ$, focusing on the Planck 2018 CMB maps. We extend the model of Preece & Battye (2014) of Gaussian correlated $Q$ and $U$ Stokes parameters to allow nonzero means. When the variances of $Q$ and $U$ are equal and their covariance and means are zero, the polarization angle is uniformly distributed. Otherwise the uniform distribution will be modulated by harmonics with $2ψ$ and $4ψ$ phases. These modulations are visible in the Planck 2018 CMB maps. Furthermore, the mean value of $U$ is peculiar compared to the power spectrum.

astro-ph.CO↗