Models, code, and papers for "Xin Lu":

Stochastic Stepwise Ensembles for Variable Selection

Mar 02, 2011
Lu Xin, Mu Zhu

In this article, we advocate the ensemble approach for variable selection. We point out that the stochastic mechanism used to generate the variable-selection ensemble (VSE) must be picked with care. We construct a VSE using a stochastic stepwise algorithm, and compare its performance with numerous state-of-the-art algorithms.

* Journal of Computational and Graphical Statistics, June 2012, Vol. 21, No. 2, Pages 275 - 294 

  Click for Model/Code and Paper
Informative Gene Selection for Microarray Classification via Adaptive Elastic Net with Conditional Mutual Information

Jun 13, 2018
Xin-Guang Yang, Yongjin Lu

Due to the advantage of achieving a better performance under weak regularization, elastic net has attracted wide attention in statistics, machine learning, bioinformatics, and other fields. In particular, a variation of the elastic net, adaptive elastic net (AEN), integrates the adaptive grouping effect. In this paper, we aim to develop a new algorithm: Adaptive Elastic Net with Conditional Mutual Information (AEN-CMI) that further improves AEN by incorporating conditional mutual information into the gene selection process. We apply this new algorithm to screen significant genes for two kinds of cancers: colon cancer and leukemia. Compared with other algorithms including Support Vector Machine, Classic Elastic Net and Adaptive Elastic Net, the proposed algorithm, AEN-CMI, obtains the best classification performance using the least number of genes.


  Click for Model/Code and Paper
Kernel Additive Principal Components

Nov 21, 2015
Xin Lu Tan, Andreas Buja, Zongming Ma

Additive principal components (APCs for short) are a nonlinear generalization of linear principal components. We focus on smallest APCs to describe additive nonlinear constraints that are approximately satisfied by the data. Thus APCs fit data with implicit equations that treat the variables symmetrically, as opposed to regression analyses which fit data with explicit equations that treat the data asymmetrically by singling out a response variable. We propose a regularized data-analytic procedure for APC estimation using kernel methods. In contrast to existing approaches to APCs that are based on regularization through subspace restriction, kernel methods achieve regularization through shrinkage and therefore grant distinctive flexibility in APC estimation by allowing the use of infinite-dimensional functions spaces for searching APC transformation while retaining computational feasibility. To connect population APCs and kernelized finite-sample APCs, we study kernelized population APCs and their associated eigenproblems, which eventually lead to the establishment of consistency of the estimated APCs. Lastly, we discuss an iterative algorithm for computing kernelized finite-sample APCs.

* 54 pages including appendices 

  Click for Model/Code and Paper
Good, Better, Best: Textual Distractors Generation for Multi-Choice VQA via Policy Gradient

Oct 21, 2019
Jiaying Lu, Xin Ye, Yi Ren, Yezhou Yang

Textual distractors in current multi-choice VQA datasets are not challenging enough for state-of-the-art neural models. To better assess whether well-trained VQA models are vulnerable to potential attack such as more challenging distractors, we introduce a novel task called \textit{textual Distractors Generation for VQA} (DG-VQA). The goal of DG-VQA is to generate the most confusing distractors in multi-choice VQA tasks represented as a tuple of image, question, and the correct answer. Consequently, such distractors expose the vulnerability of neural models. We show that distractor generation can be formulated as a Markov Decision Process, and present a reinforcement learning solution to unsupervised produce distractors. Our solution addresses the lack of large annotated corpus issue in classical distractor generation methods. Our proposed model receives reward signals from well-trained multi-choice VQA models and updates its parameters via policy gradient. The empirical results show that the generated textual distractors can successfully confuse several cutting-edge models with an average 20% accuracy drop from around 64%. Furthermore, we conduct extra adversarial training to improve the robustness of VQA models by incorporating the generated distractors. The experiment validates the effectiveness of adversarial training by showing a performance improvement of 27% for the multi-choice VQA task


  Click for Model/Code and Paper
Competitive Co-evolution for Dynamic Constrained Optimisation

Jul 31, 2019
Xiaofen Lu, Ke Tang, Stefan Menzel, Xin Yao

Dynamic constrained optimisation problems (DCOPs) widely exist in the real world due to frequently changing factors influencing the environment. Many dynamic optimisation methods such as diversity-driven methods, memory and prediction methods offer different strategies to deal with environmental changes. However, when DCOPs change very fast or have very limited time for the algorithm to react, the potential of these methods is limited due to time shortage for re-optimisation and adaptation. This is especially true for population-based dynamic optimisation methods, which normally need quite a few fitness evaluations to find a near-optimum solution. To address this issue, this paper proposes to tackle fast-changing DCOPs through a smart combination of offline and online optimisation. The offline optimisation aims to prepare a set of good solutions for all possible environmental changes beforehand. With this solution set, the online optimisation aims to react quickly to each truly happening environmental change by doing optimisation on the set. To find this solution set, this paper further proposes to use competitive co-evolution for offline optimisation by co-evolving candidate solutions and environmental parameters. The experimental studies on a well-known benchmark test set of DCOPs show that the proposed method outperforms existing methods significantly especially when the environment changes very fast


  Click for Model/Code and Paper
Grid R-CNN Plus: Faster and Better

Jun 13, 2019
Xin Lu, Buyu Li, Yuxin Yue, Quanquan Li, Junjie Yan

Grid R-CNN is a well-performed objection detection framework. It transforms the traditional box offset regression problem into a grid point estimation problem. With the guidance of the grid points, it can obtain high-quality localization results. However, the speed of Grid R-CNN is not so satisfactory. In this technical report we present Grid R-CNN Plus, a better and faster version of Grid R-CNN. We have made several updates that significantly speed up the framework and simultaneously improve the accuracy. On COCO dataset, the Res50-FPN based Grid R-CNN Plus detector achieves an mAP of 40.4%, outperforming the baseline on the same model by 3.0 points with similar inference time. Code is available at https://github.com/STVIR/Grid-R-CNN .


  Click for Model/Code and Paper
Grid R-CNN

Nov 29, 2018
Xin Lu, Buyu Li, Yuxin Yue, Quanquan Li, Junjie Yan

This paper proposes a novel object detection framework named Grid R-CNN, which adopts a grid guided localization mechanism for accurate object detection. Different from the traditional regression based methods, the Grid R-CNN captures the spatial information explicitly and enjoys the position sensitive property of fully convolutional architecture. Instead of using only two independent points, we design a multi-point supervision formulation to encode more clues in order to reduce the impact of inaccurate prediction of specific points. To take the full advantage of the correlation of points in a grid, we propose a two-stage information fusion strategy to fuse feature maps of neighbor grid points. The grid guided localization approach is easy to be extended to different state-of-the-art detection frameworks. Grid R-CNN leads to high quality object localization, and experiments demonstrate that it achieves a 4.1% AP gain at IoU=0.8 and a 10.0% AP gain at IoU=0.9 on COCO benchmark compared to Faster R-CNN with Res50 backbone and FPN architecture.


  Click for Model/Code and Paper
AxTrain: Hardware-Oriented Neural Network Training for Approximate Inference

May 21, 2018
Xin He, Liu Ke, Wenyan Lu, Guihai Yan, Xuan Zhang

The intrinsic error tolerance of neural network (NN) makes approximate computing a promising technique to improve the energy efficiency of NN inference. Conventional approximate computing focuses on balancing the efficiency-accuracy trade-off for existing pre-trained networks, which can lead to suboptimal solutions. In this paper, we propose AxTrain, a hardware-oriented training framework to facilitate approximate computing for NN inference. Specifically, AxTrain leverages the synergy between two orthogonal methods---one actively searches for a network parameters distribution with high error tolerance, and the other passively learns resilient weights by numerically incorporating the noise distributions of the approximate hardware in the forward pass during the training phase. Experimental results from various datasets with near-threshold computing and approximation multiplication strategies demonstrate AxTrain's ability to obtain resilient neural network parameters and system energy efficiency improvement.

* In International Symposium on Low Power Electronics and Design (ISLPED) 2018 

  Click for Model/Code and Paper
Rethinking the Smaller-Norm-Less-Informative Assumption in Channel Pruning of Convolution Layers

Feb 02, 2018
Jianbo Ye, Xin Lu, Zhe Lin, James Z. Wang

Model pruning has become a useful technique that improves the computational efficiency of deep learning, making it possible to deploy solutions in resource-limited scenarios. A widely-used practice in relevant work assumes that a smaller-norm parameter or feature plays a less informative role at the inference time. In this paper, we propose a channel pruning technique for accelerating the computations of deep convolutional neural networks (CNNs) that does not critically rely on this assumption. Instead, it focuses on direct simplification of the channel-to-channel computation graph of a CNN without the need of performing a computationally difficult and not-always-useful task of making high-dimensional tensors of CNN structured sparse. Our approach takes two stages: first to adopt an end-to- end stochastic training method that eventually forces the outputs of some channels to be constant, and then to prune those constant channels from the original neural network by adjusting the biases of their impacting layers such that the resulting compact model can be quickly fine-tuned. Our approach is mathematically appealing from an optimization perspective and easy to reproduce. We experimented our approach through several image learning benchmarks and demonstrate its interesting aspects and competitive performance.

* accepted to ICLR 2018, 11 pages 

  Click for Model/Code and Paper
Data Sanity Check for Deep Learning Systems via Learnt Assertions

Sep 28, 2019
Haochuan Lu, Huanlin Xu, Nana Liu, Yangfan Zhou, Xin Wang

Reliability is a critical consideration to DL-based systems. But the statistical nature of DL makes it quite vulnerable to invalid inputs, i.e., those cases that are not considered in the training phase of a DL model. This paper proposes to perform data sanity check to identify invalid inputs, so as to enhance the reliability of DL-based systems. We design and implement a tool to detect behavior deviation of a DL model when processing an input case. This tool extracts the data flow footprints and conducts an assertion-based validation mechanism. The assertions are built automatically, which are specifically-tailored for DL model data flow analysis. Our experiments conducted with real-world scenarios demonstrate that such an assertion-based data sanity check mechanism is effective in identifying invalid input cases.


  Click for Model/Code and Paper
Tetris: Re-architecting Convolutional Neural Network Computation for Machine Learning Accelerators

Nov 14, 2018
Hang Lu, Xin Wei, Ning Lin, Guihai Yan, and Xiaowei Li

Inference efficiency is the predominant consideration in designing deep learning accelerators. Previous work mainly focuses on skipping zero values to deal with remarkable ineffectual computation, while zero bits in non-zero values, as another major source of ineffectual computation, is often ignored. The reason lies on the difficulty of extracting essential bits during operating multiply-and-accumulate (MAC) in the processing element. Based on the fact that zero bits occupy as high as 68.9% fraction in the overall weights of modern deep convolutional neural network models, this paper firstly proposes a weight kneading technique that could eliminate ineffectual computation caused by either zero value weights or zero bits in non-zero weights, simultaneously. Besides, a split-and-accumulate (SAC) computing pattern in replacement of conventional MAC, as well as the corresponding hardware accelerator design called Tetris are proposed to support weight kneading at the hardware level. Experimental results prove that Tetris could speed up inference up to 1.50x, and improve power efficiency up to 5.33x compared with the state-of-the-art baselines.

* ICCAD 2018 paper 

  Click for Model/Code and Paper
Large-scale Land Cover Classification in GaoFen-2 Satellite Imagery

Jun 04, 2018
Xin-Yi Tong, Qikai Lu, Gui-Song Xia, Liangpei Zhang

Many significant applications need land cover information of remote sensing images that are acquired from different areas and times, such as change detection and disaster monitoring. However, it is difficult to find a generic land cover classification scheme for different remote sensing images due to the spectral shift caused by diverse acquisition condition. In this paper, we develop a novel land cover classification method that can deal with large-scale data captured from widely distributed areas and different times. Additionally, we establish a large-scale land cover classification dataset consisting of 150 Gaofen-2 imageries as data support for model training and performance evaluation. Our experiments achieve outstanding classification accuracy compared with traditional methods.

* IGARSS'18 conference paper 

  Click for Model/Code and Paper
Neural Generative Question Answering

Apr 22, 2016
Jun Yin, Xin Jiang, Zhengdong Lu, Lifeng Shang, Hang Li, Xiaoming Li

This paper presents an end-to-end neural network model, named Neural Generative Question Answering (GENQA), that can generate answers to simple factoid questions, based on the facts in a knowledge-base. More specifically, the model is built on the encoder-decoder framework for sequence-to-sequence learning, while equipped with the ability to enquire the knowledge-base, and is trained on a corpus of question-answer pairs, with their associated triples in the knowledge-base. Empirical study shows the proposed model can effectively deal with the variations of questions and answers, and generate right and natural answers by referring to the facts in the knowledge-base. The experiment on question answering demonstrates that the proposed model can outperform an embedding-based QA model as well as a neural dialogue model trained on the same data.

* Accepted by IJCAI 2016 

  Click for Model/Code and Paper
Focus-Enhanced Scene Text Recognition with Deformable Convolutions

Sep 23, 2019
Linjie Deng, Yanxiang Gong, Xinchen Lu, Xin Yi, Zheng Ma, Mei Xie

Recently, scene text recognition methods based on deep learning have sprung up in computer vision area. The existing methods achieved great performances, but the recognition of irregular text is still challenging due to the various shapes and distorted patterns. Consider that at the time of reading words in the real world, normally we will not rectify it in our mind but adjust our focus and visual fields. Similarly, through utilizing deformable convolutional layers whose geometric structures are adjustable, we present an enhanced recognition network without the steps of rectification to deal with irregular text in this work. A number of experiments have been applied, where the results on public benchmarks demonstrate the effectiveness of our proposed components and shows that our method has reached satisfactory performances. The code will be publicly available at https://github.com/Alpaca07/dtr soon.


  Click for Model/Code and Paper
FVNet: 3D Front-View Proposal Generation for Real-Time Object Detection from Point Clouds

Mar 26, 2019
Jie Zhou, Xuequan Lu, Xin Tan, Zhiwei Shao, Shouhong Ding, Lizhuang Ma

3D object detection from raw and sparse point clouds has been far less treated to date, compared with its 2D counterpart. In this paper, we propose a novel framework called FVNet for 3D front-view proposal generation and object detection from point clouds. It consists of two stages: generation of front-view proposals and estimation of 3D bounding box parameters. Instead of generating proposals from camera images or bird's-eye-view maps, we first project point clouds onto a cylindrical surface to generate front-view feature maps which retains rich information. We then introduce a proposal generation network to predict 3D region proposals from the generated maps and further extrude objects of interest from the whole point cloud. Finally, we present another network to extract the point-wise features from the extruded object points and regress the final 3D bounding box parameters in the canonical coordinates. Our framework achieves real-time performance with 12ms per point cloud sample. Extensive experiments on the 3D detection benchmark KITTI show that the proposed architecture outperforms state-of-the-art techniques which take either camera images or point clouds as input, in terms of accuracy and inference time.

* 10 pages, 6 figures 

  Click for Model/Code and Paper
The Newton Scheme for Deep Learning

Oct 16, 2018
Junqing Qiu, Guoren Zhong, Yihua Lu, Kun Xin, Huihuan Qian, Xi Zhu

We introduce a neural network (NN) strictly governed by Newton's Law, with the nature required basis functions derived from the fundamental classic mechanics. Then, by classifying the training model as a quick procedure of 'force pattern' recognition, we developed the Newton physics-based NS scheme. Once the force pattern is confirmed, the neuro network simply does the checking of the 'pattern stability' instead of the continuous fitting by computational resource consuming big data-driven processing. In the given physics's law system, once the field is confirmed, the mathematics bases for the force field description actually are not diverged but denumerable, which can save the function representations from the exhaustible available mathematics bases. In this work, we endorsed Newton's Law into the deep learning technology and proposed Newton Scheme (NS). Under NS, the user first identifies the path pattern, like the constant acceleration movement.The object recognition technology first loads mass information, then, the NS finds the matched physical pattern and describe and predict the trajectory of the movements with nearly zero error. We compare the major contribution of this NS with the TCN, GRU and other physics inspired 'FIND-PDE' methods to demonstrate fundamental and extended applications of how the NS works for the free-falling, pendulum and curve soccer balls.The NS methodology provides more opportunity for the future deep learning advances.

* 7 pages, 10 figures 

  Click for Model/Code and Paper
Flow-Grounded Spatial-Temporal Video Prediction from Still Images

Aug 26, 2018
Yijun Li, Chen Fang, Jimei Yang, Zhaowen Wang, Xin Lu, Ming-Hsuan Yang

Existing video prediction methods mainly rely on observing multiple historical frames or focus on predicting the next one-frame. In this work, we study the problem of generating consecutive multiple future frames by observing one single still image only. We formulate the multi-frame prediction task as a multiple time step flow (multi-flow) prediction phase followed by a flow-to-frame synthesis phase. The multi-flow prediction is modeled in a variational probabilistic manner with spatial-temporal relationships learned through 3D convolutions. The flow-to-frame synthesis is modeled as a generative process in order to keep the predicted results lying closer to the manifold shape of real video sequence. Such a two-phase design prevents the model from directly looking at the high-dimensional pixel space of the frame sequence and is demonstrated to be more effective in predicting better and diverse results. Extensive experimental results on videos with different types of motion show that the proposed algorithm performs favorably against existing methods in terms of quality, diversity and human perceptual evaluation.

* Accepted by ECCV 2018 

  Click for Model/Code and Paper
Universal Style Transfer via Feature Transforms

Nov 17, 2017
Yijun Li, Chen Fang, Jimei Yang, Zhaowen Wang, Xin Lu, Ming-Hsuan Yang

Universal style transfer aims to transfer arbitrary visual styles to content images. Existing feed-forward based methods, while enjoying the inference efficiency, are mainly limited by inability of generalizing to unseen styles or compromised visual quality. In this paper, we present a simple yet effective method that tackles these limitations without training on any pre-defined styles. The key ingredient of our method is a pair of feature transforms, whitening and coloring, that are embedded to an image reconstruction network. The whitening and coloring transforms reflect a direct matching of feature covariance of the content image to a given style image, which shares similar spirits with the optimization of Gram matrix based cost in neural style transfer. We demonstrate the effectiveness of our algorithm by generating high-quality stylized images with comparisons to a number of recent methods. We also analyze our method by visualizing the whitened features and synthesizing textures via simple feature coloring.

* Accepted by NIPS 2017 

  Click for Model/Code and Paper