arXiv subjects

Haoran Li

Publications and source records attributed to Haoran Li.

334 records · Page 19Linked to original sources

arXiv2020

General Purpose Text Embeddings from Pre-trained Language Models for Scalable Inference

The state of the art on many NLP tasks is currently achieved by large pre-trained language models, which require a considerable amount of computation. We explore a setting where many different predictions are made on a single piece of text. In that case, some of the computational cost during inference can be amortized over the different tasks using a shared text encoder. We compare approaches for training such an encoder and show that encoders pre-trained over multiple tasks generalize well to unseen tasks. We also compare ways of extracting fixed- and limited-size representations from this encoder, including different ways of pooling features extracted from multiple layers or positions. Our best approach compares favorably to knowledge distillation, achieving higher accuracy and lower computational cost once the system is handling around 7 tasks. Further, we show that through binary quantization, we can reduce the size of the extracted representations by a factor of 16 making it feasible to store them for later use. The resulting method offers a compelling solution for using large-scale pre-trained models at a fraction of the computational cost when multiple tasks are performed on the same text.

cs.CL↗

arXiv2020

BiFNet: Bidirectional Fusion Network for Road Segmentation

Multi-sensor fusion-based road segmentation plays an important role in the intelligent driving system since it provides a drivable area. The existing mainstream fusion method is mainly to feature fusion in the image space domain which causes the perspective compression of the road and damages the performance of the distant road. Considering the bird's eye views(BEV) of the LiDAR remains the space structure in horizontal plane, this paper proposes a bidirectional fusion network(BiFNet) to fuse the image and BEV of the point cloud. The network consists of two modules: 1) Dense space transformation module, which solves the mutual conversion between camera image space and BEV space. 2) Context-based feature fusion module, which fuses the different sensors information based on the scenes from corresponding features.This method has achieved competitive results on KITTI dataset.

cs.CV↗

arXiv2020

Multi-modal Datasets for Super-resolution

Nowdays, most datasets used to train and evaluate super-resolution models are single-modal simulation datasets. However, due to the variety of image degradation types in the real world, models trained on single-modal simulation datasets do not always have good robustness and generalization ability in different degradation scenarios. Previous work tended to focus only on true-color images. In contrast, we first proposed real-world black-and-white old photo datasets for super-resolution (OID-RW), which is constructed using two methods of manually filling pixels and shooting with different cameras. The dataset contains 82 groups of images, including 22 groups of character type and 60 groups of landscape and architecture. At the same time, we also propose a multi-modal degradation dataset (MDD400) to solve the super-resolution reconstruction in real-life image degradation scenarios. We managed to simulate the process of generating degraded images by the following four methods: interpolation algorithm, CNN network, GAN network and capturing videos with different bit rates. Our experiments demonstrate that not only the models trained on our dataset have better generalization capability and robustness, but also the trained images can maintain better edge contours and texture features.

eess.IV↗

arXiv2019

Optimal Multi-View Video Transmission in OFDMA Systems

In this letter, we study the transmission of a multi-view video (MVV) to multiple users in an Orthogonal Frequency Division Multiple Access (OFDMA) system. To maximally improve transmission efficiency, we exploit both natural multicast opportunities and view synthesis-enabled multicast opportunities. First, we establish a communication model for transmission of a MVV to multiple users in an OFDMA system. Then, we formulate the minimization problem of the average weighted sum energy consumption for view transmission and synthesis with respect to view selection and transmission power and subcarrier allocation. The optimization problem is a challenging mixed discrete-continuous optimization problem with huge numbers of variables and constraints. A low-complexity algorithm is proposed to obtain a suboptimal solution. Finally, numerical results further demonstrate the value of view synthesis-enabled multicast opportunities for MVV transmission in OFDMA systems.

cs.IT↗

arXiv2019

Music-oriented Dance Video Synthesis with Pose Perceptual Loss

We present a learning-based approach with pose perceptual loss for automatic music video generation. Our method can produce a realistic dance video that conforms to the beats and rhymes of almost any given music. To achieve this, we firstly generate a human skeleton sequence from music and then apply the learned pose-to-appearance mapping to generate the final video. In the stage of generating skeleton sequences, we utilize two discriminators to capture different aspects of the sequence and propose a novel pose perceptual loss to produce natural dances. Besides, we also provide a new cross-modal evaluation to evaluate the dance quality, which is able to estimate the similarity between two modalities of music and dance. Finally, a user study is conducted to demonstrate that dance video synthesized by the presented approach produces surprisingly realistic results. The results are shown in the supplementary video at https://youtu.be/0rMuFMZa_K4

cs.CV↗

arXiv2019

Predictive Multi-level Patient Representations from Electronic Health Records

The advent of the Internet era has led to an explosive growth in the Electronic Health Records (EHR) in the past decades. The EHR data can be regarded as a collection of clinical events, including laboratory results, medication records, physiological indicators, etc, which can be used for clinical outcome prediction tasks to support constructions of intelligent health systems. Learning patient representation from these clinical events for the clinical outcome prediction is an important but challenging step. Most related studies transform EHR data of a patient into a sequence of clinical events in temporal order and then use sequential models to learn patient representations for outcome prediction. However, clinical event sequence contains thousands of event types and temporal dependencies. We further make an observation that clinical events occurring in a short period are not constrained by any temporal order but events in a long term are influenced by temporal dependencies. The multi-scale temporal property makes it difficult for traditional sequential models to capture the short-term co-occurrence and the long-term temporal dependencies in clinical event sequences. In response to the above challenges, this paper proposes a Multi-level Representation Model (MRM). MRM first uses a sparse attention mechanism to model the short-term co-occurrence, then uses interval-based event pooling to remove redundant information and reduce sequence length and finally predicts clinical outcomes through Long Short-Term Memory (LSTM). Experiments on real-world datasets indicate that our proposed model largely improves the performance of clinical outcome prediction tasks using EHR data.

cs.LG↗

arXiv2019

Learning Hierarchical Representations of Electronic Health Records for Clinical Outcome Prediction

Clinical outcome prediction based on the Electronic Health Record (EHR) plays a crucial role in improving the quality of healthcare. Conventional deep sequential models fail to capture the rich temporal patterns encoded in the longand irregular clinical event sequences. We make the observation that clinical events at a long time scale exhibit strongtemporal patterns, while events within a short time period tend to be disordered co-occurrence. We thus propose differentiated mechanisms to model clinical events at different time scales. Our model learns hierarchical representationsof event sequences, to adaptively distinguish between short-range and long-range events, and accurately capture coretemporal dependencies. Experimental results on real clinical data show that our model greatly improves over previous state-of-the-art models, achieving AUC scores of 0.94 and 0.90 for predicting death and ICU admission respectively, Our model also successfully identifies important events for different clinical outcome prediction tasks

cs.LG↗

arXiv2019

Robust Hidden Topology Identification in Distribution Systems

With more distributed energy resources (DERs) connected to distribution grids, better monitoring and control are needed, where identifying the topology accurately is the prerequisite. However, due to frequent re-configurations, operators usually cannot know a complete structure in distribution grids. Luckily, the growing data from smart sensors, restricted by Ohm law, provides the possibility of topology inference. In this paper, we show how line parameters of Ohm equation can be estimated for topology identification even when there are hidden nodes. Specifically, the introduced learning method recursively conducts hidden-node detection and impedance calculation. However, the assumptions on uncorrelated data, availability of phasor measurements, and a balanced system, are not met in practices, causing large errors. To resolve these problems, we employ Cholesky whitening first with a proof for measurement decorrelations. For increasing robustness further, we show how to handle practical scenarios when only measurement magnitudes are available or when the grid is three-phase unbalanced. Numerical performance is verified on multi-size distribution grids with both simulation and real-world data.

eess.SY↗

arXiv2018

High-dimensional general linear hypothesis tests via non-linear spectral shrinkage

We are interested in testing general linear hypotheses in a high-dimensional multivariate linear regression model. The framework includes many well-studied problems such as two-sample tests for equality of population means, MANOVA and others as special cases. A family of rotation-invariant tests is proposed that involves a flexible spectral shrinkage scheme applied to the sample error covariance matrix. The asymptotic normality of the test statistic under the null hypothesis is derived in the setting where dimensionality is comparable to sample sizes, assuming the existence of certain moments for the observations. The asymptotic power of the proposed test is studied under various local alternatives. The power characteristics are then utilized to propose a data-driven selection of the spectral shrinkage function. As an illustration of the general theory, we construct a family of tests involving ridge-type regularization and suggest possible extensions to more complex regularizers. A simulation study is carried out to examine the numerical performance of the proposed tests.

stat.ME↗

arXiv2018

An adaptable generalization of Hotelling's $T^2$ test in high dimension

We propose a two-sample test for detecting the difference between mean vectors in a high-dimensional regime based on a ridge-regularized Hotelling's $T^2$. To choose the regularization parameter, a method is derived that aims at maximizing power within a class of local alternatives. We also propose a composite test that combines the optimal tests corresponding to a specific collection of local alternatives. Weak convergence of the stochastic process corresponding to the ridge-regularized Hotelling's $T^2$ is established and used to derive the cut-off values of the proposed test. Large sample properties are verified for a class of sub-Gaussian distributions. Through an extensive simulation study, the composite test is shown to compare favorably against a host of existing two-sample test procedure in a wide range of settings. The performance of the proposed test procedure is illustrated through an application to a breast cancer data set where the goal is to detect the pathways with different DNA copy number alterations across breast cancer subtypes.

stat.ME↗