arXiv subjects

Ze Wang

Publications and source records attributed to Ze Wang.

138 records · Page 8Linked to original sources

arXiv2020

Assessing the neurocognitive correlates of resting brain entropy

The human brain exhibits large-scale spontaneous fluctuations that account for most of its total energy metabolism. Independent of any overt function, this immense ongoing activity likely creates or maintains a potential functional brain reserve to facilitate normal brain function. An important property of spontaneous brain activity is the long-range temporal coherence, which can be characterized by resting state fMRI-based brain entropy mapping (BEN), a relatively new method that has gained increasing research interest. The purpose of this study was to leverage the large resting state fMRI and behavioral data publicly available from the human connectome project to address three important but still unknown questions: temporal stability of rsfMRI-derived BEN; the relationship of resting BEN to latent functional reserve; associations of resting BEN to neurocognition. Our results showed that rsfMRI-derived BEN was highly stable across time; resting BEN in the default mode network (DMN) and executive control network (ECN) was related to brain reserve in a negative correlation to education years; and lower DMN/ECN BEN corresponds to higher fluid intelligence and better task performance. These results suggest that resting BEN is a temporally stable brain trait; BEN in DMN/ECN may provide a means to measure the latent functional reserve that bestows better brain functionality and may be enhanced by education.

q-bio.NC↗

arXiv2020

Stochastic Conditional Generative Networks with Basis Decomposition

While generative adversarial networks (GANs) have revolutionized machine learning, a number of open questions remain to fully understand them and exploit their power. One of these questions is how to efficiently achieve proper diversity and sampling of the multi-mode data space. To address this, we introduce BasisGAN, a stochastic conditional multi-mode image generator. By exploiting the observation that a convolutional filter can be well approximated as a linear combination of a small set of basis elements, we learn a plug-and-played basis generator to stochastically generate basis elements, with just a few hundred of parameters, to fully embed stochasticity into convolutional filters. By sampling basis elements instead of filters, we dramatically reduce the cost of modeling the parameter space with no sacrifice on either image diversity or fidelity. To illustrate this proposed plug-and-play framework, we construct variants of BasisGAN based on state-of-the-art conditional image generation networks, and train the networks by simply plugging in a basis generator, without additional auxiliary components, hyperparameters, or training objectives. The experimental success is complemented with theoretical results indicating how the perturbations introduced by the proposed sampling of basis elements can propagate to the appearance of generated images.

cs.CV↗

arXiv2019

Range Adaptation for 3D Object Detection in LiDAR

LiDAR-based 3D object detection plays a crucial role in modern autonomous driving systems. LiDAR data often exhibit severe changes in properties across different observation ranges. In this paper, we explore cross-range adaptation for 3D object detection using LiDAR, i.e., far-range observations are adapted to near-range. This way, far-range detection is optimized for similar performance to near-range one. We adopt a bird-eyes view (BEV) detection framework to perform the proposed model adaptation. Our model adaptation consists of an adversarial global adaptation, and a fine-grained local adaptation. The proposed cross range adaptation framework is validated on three state-of-the-art LiDAR based object detection networks, and we consistently observe performance improvement on the far-range objects, without adding any auxiliary parameters to the model. To the best of our knowledge, this paper is the first attempt to study cross-range LiDAR adaptation for object detection in point clouds. To demonstrate the generality of the proposed adaptation framework, experiments on more challenging cross-device adaptation are further conducted, and a new LiDAR dataset with high-quality annotated point clouds is released to promote future research.

cs.CV↗

arXiv2019

The Structure Transfer Machine Theory and Applications

Representation learning is a fundamental but challenging problem, especially when the distribution of data is unknown. We propose a new representation learning method, termed Structure Transfer Machine (STM), which enables feature learning process to converge at the representation expectation in a probabilistic way. We theoretically show that such an expected value of the representation (mean) is achievable if the manifold structure can be transferred from the data space to the feature space. The resulting structure regularization term, named manifold loss, is incorporated into the loss function of the typical deep learning pipeline. The STM architecture is constructed to enforce the learned deep representation to satisfy the intrinsic manifold structure from the data, which results in robust features that suit various application scenarios, such as digit recognition, image classification and object tracking. Compared to state-of-the-art CNN architectures, we achieve the better results on several commonly used benchmarks\footnote{The source code is available. https://github.com/stmstmstm/stm }.

cs.LG↗

arXiv2019

Understanding Urban Dynamics via Context-aware Tensor Factorization with Neighboring Regularization

Recent years have witnessed the world-wide emergence of mega-metropolises with incredibly huge populations. Understanding residents mobility patterns, or urban dynamics, thus becomes crucial for building modern smart cities. In this paper, we propose a Neighbor-Regularized and context-aware Non-negative Tensor Factorization model (NR-cNTF) to discover interpretable urban dynamics from urban heterogeneous data. Different from many existing studies concerned with prediction tasks via tensor completion, NR-cNTF focuses on gaining urban managerial insights from spatial, temporal, and spatio-temporal patterns. This is enabled by high-quality Tucker factorizations regularized by both POI-based urban contexts and geographically neighboring relations. NR-cNTF is also capable of unveiling long-term evolutions of urban dynamics via a pipeline initialization approach. We apply NR-cNTF to a real-life data set containing rich taxi GPS trajectories and POI records of Beijing. The results indicate: 1) NR-cNTF accurately captures four kinds of city rhythms and seventeen spatial communities; 2) the rapid development of Beijing, epitomized by the CBD area, indeed intensifies the job-housing imbalance; 3) the southern areas with recent government investments have shown more healthy development tendency. Finally, NR-cNTF is compared with some baselines on traffic prediction, which further justifies the importance of urban contexts awareness and neighboring regulations.

cs.LG↗

arXiv2018

CD-CNN: A Partially Supervised Cross-Domain Deep Learning Model for Urban Resident Recognition

Driven by the wave of urbanization in recent decades, the research topic about migrant behavior analysis draws great attention from both academia and the government. Nevertheless, subject to the cost of data collection and the lack of modeling methods, most of existing studies use only questionnaire surveys with sparse samples and non-individual level statistical data to achieve coarse-grained studies of migrant behaviors. In this paper, a partially supervised cross-domain deep learning model named CD-CNN is proposed for migrant/native recognition using mobile phone signaling data as behavioral features and questionnaire survey data as incomplete labels. Specifically, CD-CNN features in decomposing the mobile data into location domain and communication domain, and adopts a joint learning framework that combines two convolutional neural networks with a feature balancing scheme. Moreover, CD-CNN employs a three-step algorithm for training, in which the co-training step is of great value to partially supervised cross-domain learning. Comparative experiments on the city Wuxi demonstrate the high predictive power of CD-CNN. Two interesting applications further highlight the ability of CD-CNN for in-depth migrant behavioral analysis.

cs.CY↗

arXiv2018

In Defense of Single-column Networks for Crowd Counting

Crowd counting usually addressed by density estimation becomes an increasingly important topic in computer vision due to its widespread applications in video surveillance, urban planning, and intelligence gathering. However, it is essentially a challenging task because of the greatly varied sizes of objects, coupled with severe occlusions and vague appearance of extremely small individuals. Existing methods heavily rely on multi-column learning architectures to extract multi-scale features, which however suffer from heavy computational cost, especially undesired for crowd counting. In this paper, we propose the single-column counting network (SCNet) for efficient crowd counting without relying on multi-column networks. SCNet consists of residual fusion modules (RFMs) for multi-scale feature extraction, a pyramid pooling module (PPM) for information fusion, and a sub-pixel convolutional module (SPCM) followed by a bilinear upsampling layer for resolution recovery. Those proposed modules enable our SCNet to fully capture multi-scale features in a compact single-column architecture and estimate high-resolution density map in an efficient way. In addition, we provide a principled paradigm for density map generation and data augmentation for training, which shows further improved performance. Extensive experiments on three benchmark datasets show that our SCNet delivers new state-of-the-art performance and surpasses previous methods by large margins, which demonstrates the great effectiveness of SCNet as a single-column network for crowd counting.

cs.CV↗

arXiv2018

LaneNet: Real-Time Lane Detection Networks for Autonomous Driving

Lane detection is to detect lanes on the road and provide the accurate location and shape of each lane. It severs as one of the key techniques to enable modern assisted and autonomous driving systems. However, several unique properties of lanes challenge the detection methods. The lack of distinctive features makes lane detection algorithms tend to be confused by other objects with similar local appearance. Moreover, the inconsistent number of lanes on a road as well as diverse lane line patterns, e.g. solid, broken, single, double, merging, and splitting lines further hamper the performance. In this paper, we propose a deep neural network based method, named LaneNet, to break down the lane detection into two stages: lane edge proposal and lane line localization. Stage one uses a lane edge proposal network for pixel-wise lane edge classification, and the lane line localization network in stage two then detects lane lines based on lane edge proposals. Please note that the goal of our LaneNet is built to detect lane line only, which introduces more difficulties on suppressing the false detections on the similar lane marks on the road like arrows and characters. Despite all the difficulties, our lane detection is shown to be robust to both highway and urban road scenarios method without relying on any assumptions on the lane number or the lane line patterns. The high running speed and low computational cost endow our LaneNet the capability of being deployed on vehicle-based systems. Experiments validate that our LaneNet consistently delivers outstanding performances on real world traffic scenarios.

cs.CV↗

arXiv2018

Multilevel Wavelet Decomposition Network for Interpretable Time Series Analysis

Recent years have witnessed the unprecedented rising of time series from almost all kindes of academic and industrial fields. Various types of deep neural network models have been introduced to time series analysis, but the important frequency information is yet lack of effective modeling. In light of this, in this paper we propose a wavelet-based neural network structure called multilevel Wavelet Decomposition Network (mWDN) for building frequency-aware deep learning models for time series analysis. mWDN preserves the advantage of multilevel discrete wavelet decomposition in frequency learning while enables the fine-tuning of all parameters under a deep neural network framework. Based on mWDN, we further propose two deep learning models called Residual Classification Flow (RCF) and multi-frequecy Long Short-Term Memory (mLSTM) for time series classification and forecasting, respectively. The two models take all or partial mWDN decomposed sub-series in different frequencies as input, and resort to the back propagation algorithm to learn all the parameters globally, which enables seamless embedding of wavelet-based frequency analysis into deep learning frameworks. Extensive experiments on 40 UCR datasets and a real-world user volume dataset demonstrate the excellent performance of our time series models based on mWDN. In particular, we propose an importance analysis method to mWDN based models, which successfully identifies those time-series elements and mWDN layers that are crucially important to time series analysis. This indeed indicates the interpretability advantage of mWDN, and can be viewed as an indepth exploration to interpretable deep learning.

cs.LG↗

arXiv2018

Leveraging Social Signal to Improve Item Recommendation for Matrix Factorization

Although Recommender Systems have been comprehensively studied in the past decade both in industry and academia, most of current recommender systems suffer from the following issues: 1) The data sparsity of the user-item matrix seriously affect the recommender system quality. As a result, most of traditional recommender system approaches are not able to deal with the users who have rated few items, which is known as cold start problem in recommender system. 2) Traditional recommender systems assume that users are independently and identically distributed and ignore the social relation between users. However, in real life scenario, due to the exponential growth of social networking service, such as facebook and Twitter, social connections between different users play an significant role for recommender system task. In this work, aiming at providing a better recommender systems by incorporating user social network information, we propose a matrix factorization framework with user social connection constraints. Experimental results on the real-life dataset shows that the proposed method performs significantly better than the state-of-the-art approaches in terms of MAE and RMSE, especially for the cold start users.

cs.IR↗

arXiv2018

Denoising Arterial Spin Labeling Cerebral Blood Flow Images Using Deep Learning

Arterial spin labeling perfusion MRI is a noninvasive technique for measuring quantitative cerebral blood flow (CBF), but the measurement is subject to a low signal-to-noise-ratio(SNR). Various post-processing methods have been proposed to denoise ASL MRI but only provide moderate improvement. Deep learning (DL) is an emerging technique that can learn the most representative signal from data without prior modeling which can be highly complex and analytically indescribable. The purpose of this study was to assess whether the record breaking performance of DL can be translated into ASL MRI denoising. We used convolutional neural network (CNN) to build the DL ASL denosing model (DL-ASL) to inherently consider the inter-voxel correlations. To better guide DL-ASL training, we incorporated prior knowledge about ASL MRI: the structural similarity between ASL CBF map and grey matter probability map. A relatively large sample data were used to train the model which was subsequently applied to a new set of data for testing. Experimental results showed that DL-ASL achieved state-of-the-art denoising performance for ASL MRI as compared to current routine methods in terms of higher SNR, keeping CBF quantification quality while shorten the acquisition time by 75%, and automatic partial volume correction.

cs.CV↗

arXiv2015

MRF-ZOOM: A Fast Dictionary Searching Algorithm for Magnetic Resonance Fingerprinting

Magnetic resonance fingerprinting (MRF) is a new technique for simultaneously quantifying multiple MR parameters using one temporally resolved MR scan. But its brute-force dictionary generating and searching (DGS) process causes a huge disk space demand and computational burden, prohibiting it from a practical multiple slice high-definition imaging. The purpose of this paper was to provide a fast and space efficient DGS algorithm for MRF. Based on an empirical analysis of properties of the distance function of the acquired MRF signal and the pre-defined MRF dictionary entries, we proposed a parameter separable MRF DGS method, which breaks the multiplicative computation complexity into an additive one and enabling a resolution scalable multi-resolution DGS process, which was dubbed as MRF ZOOM. The evaluation results showed that MRF ZOOM was hundreds or thousands of times faster than the original brute-force DGS method. The acceleration was even higher when considering the time difference for generating the dictionary. Using a high precision quantification, MRF can find the right parameter values for a 64x64 imaging slice in 117 secs. Our data also showed that spatial constraints can be used to further speed up MRF ZOOM.

cs.DS↗