arXiv subjects

Hao Chen

Publications and source records attributed to Hao Chen.

At least 1,387 records · Page 77Linked to original sources

arXiv2018

CHET: Compiler and Runtime for Homomorphic Evaluation of Tensor Programs

Fully Homomorphic Encryption (FHE) refers to a set of encryption schemes that allow computations to be applied directly on encrypted data without requiring a secret key. This enables novel application scenarios where a client can safely offload storage and computation to a third-party cloud provider without having to trust the software and the hardware vendors with the decryption keys. Recent advances in both FHE schemes and implementations have moved such applications from theoretical possibilities into the realm of practicalities. This paper proposes a compact and well-reasoned interface called the Homomorphic Instruction Set Architecture (HISA) for developing FHE applications. Just as the hardware ISA interface enabled hardware advances to proceed independent of software advances in the compiler and language runtimes, HISA decouples compiler optimizations and runtimes for supporting FHE applications from advancements in the underlying FHE schemes. This paper demonstrates the capabilities of HISA by building an end-to-end software stack for evaluating neural network models on encrypted data. Our stack includes an end-to-end compiler, runtime, and a set of optimizations. Our approach shows generated code, on a set of popular neural network architectures, is faster than hand-optimized implementations.

cs.LG↗

arXiv2018

Disambiguating fine-grained place names from descriptions by clustering

Everyday place descriptions often contain place names of fine-grained features, such as buildings or businesses, that are more difficult to disambiguate than names referring to larger places, for example cities or natural geographic features. Fine-grained places are often significantly more frequent and more similar to each other, and disambiguation heuristics developed for larger places, such as those based on population or containment relationships, are often not applicable in these cases. In this research, we address the disambiguation of fine-grained place names from everyday place descriptions. For this purpose, we evaluate the performance of different existing clustering-based approaches, since clustering approaches require no more knowledge other than the locations of ambiguous place names. We consider not only approaches developed specifically for place name disambiguation, but also clustering algorithms developed for general data mining that could potentially be leveraged. We compare these methods with a novel algorithm, and show that the novel algorithm outperforms the other algorithms in terms of disambiguation precision and distance error over several tested datasets.

cs.IR↗

arXiv2018

Semi-supervised Skin Lesion Segmentation via Transformation Consistent Self-ensembling Model

Automatic skin lesion segmentation on dermoscopic images is an essential component in computer-aided diagnosis of melanoma. Recently, many fully supervised deep learning based methods have been proposed for automatic skin lesion segmentation. However, these approaches require massive pixel-wise annotation from experienced dermatologists, which is very costly and time-consuming. In this paper, we present a novel semi-supervised method for skin lesion segmentation by leveraging both labeled and unlabeled data. The network is optimized by the weighted combination of a common supervised loss for labeled inputs only and a regularization loss for both labeled and unlabeled data. In this paper, we present a novel semi-supervised method for skin lesion segmentation, where the network is optimized by the weighted combination of a common supervised loss for labeled inputs only and a regularization loss for both labeled and unlabeled data. Our method encourages a consistent prediction for unlabeled images using the outputs of the network-in-training under different regularizations, so that it can utilize the unlabeled data. To utilize the unlabeled data, our method encourages the consistent predictions of the network-in-training for the same input under different regularizations. Aiming for the semi-supervised segmentation problem, we enhance the effect of regularization for pixel-level predictions by introducing a transformation, including rotation and flipping, consistent scheme in our self-ensembling model. With only 300 labeled training samples, our method sets a new record on the benchmark of the International Skin Imaging Collaboration (ISIC) 2017 skin lesion segmentation challenge. Such a result clearly surpasses fully-supervised state-of-the-arts that are trained with 2000 labeled data.

cs.CV↗

arXiv2018

H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes

Liver cancer is one of the leading causes of cancer death. To assist doctors in hepatocellular carcinoma diagnosis and treatment planning, an accurate and automatic liver and tumor segmentation method is highly demanded in the clinical practice. Recently, fully convolutional neural networks (FCNs), including 2D and 3D FCNs, serve as the back-bone in many volumetric image segmentation. However, 2D convolutions can not fully leverage the spatial information along the third dimension while 3D convolutions suffer from high computational cost and GPU memory consumption. To address these issues, we propose a novel hybrid densely connected UNet (H-DenseUNet), which consists of a 2D DenseUNet for efficiently extracting intra-slice features and a 3D counterpart for hierarchically aggregating volumetric contexts under the spirit of the auto-context algorithm for liver and tumor segmentation. We formulate the learning process of H-DenseUNet in an end-to-end manner, where the intra-slice representations and inter-slice features can be jointly optimized through a hybrid feature fusion (HFF) layer. We extensively evaluated our method on the dataset of MICCAI 2017 Liver Tumor Segmentation (LiTS) Challenge and 3DIRCADb Dataset. Our method outperformed other state-of-the-arts on the segmentation results of tumors and achieved very competitive performance for liver segmentation even with a single model.

cs.CV↗

arXiv2018

Unsupervised Cross-Modality Domain Adaptation of ConvNets for Biomedical Image Segmentations with Adversarial Loss

Convolutional networks (ConvNets) have achieved great successes in various challenging vision tasks. However, the performance of ConvNets would degrade when encountering the domain shift. The domain adaptation is more significant while challenging in the field of biomedical image analysis, where cross-modality data have largely different distributions. Given that annotating the medical data is especially expensive, the supervised transfer learning approaches are not quite optimal. In this paper, we propose an unsupervised domain adaptation framework with adversarial learning for cross-modality biomedical image segmentations. Specifically, our model is based on a dilated fully convolutional network for pixel-wise prediction. Moreover, we build a plug-and-play domain adaptation module (DAM) to map the target input to features which are aligned with source domain feature space. A domain critic module (DCM) is set up for discriminating the feature space of both domains. We optimize the DAM and DCM via an adversarial loss without using any target domain label. Our proposed method is validated by adapting a ConvNet trained with MRI images to unpaired CT data for cardiac structures segmentations, and achieved very promising results.

cs.CV↗

arXiv2018

Semantic-Aware Generative Adversarial Nets for Unsupervised Domain Adaptation in Chest X-ray Segmentation

In spite of the compelling achievements that deep neural networks (DNNs) have made in medical image computing, these deep models often suffer from degraded performance when being applied to new test datasets with domain shift. In this paper, we present a novel unsupervised domain adaptation approach for segmentation tasks by designing semantic-aware generative adversarial networks (GANs). Specifically, we transform the test image into the appearance of source domain, with the semantic structural information being well preserved, which is achieved by imposing a nested adversarial learning in semantic label space. In this way, the segmentation DNN learned from the source domain is able to be directly generalized to the transformed test image, eliminating the need of training a new model for every new target dataset. Our domain adaptation procedure is unsupervised, without using any target domain labels. The adversarial learning of our network is guided by a GAN loss for mapping data distributions, a cycle-consistency loss for retaining pixel-level content, and a semantic-aware loss for enhancing structural information. We validated our method on two different chest X-ray public datasets for left/right lung segmentation. Experimental results show that the segmentation performance of our unsupervised approach is highly competitive with the upper bound of supervised transfer learning.

cs.CV↗

arXiv2018

Deformation, lattice instability, and metallization during solid-solid structural transformations under general applied stress tensor: example of Si I -> Si II

Density functional theory (DFT) was employed to study the stress-strain behavior, elastic instabilities, and metallization during a solid-solid phase transformation (PT) between semiconducting Si I (cubic A4) and metallic Si II (tetragonal A5 structure) when subjected to a general stress tensor. With normal stresses ($σ_1$, $σ_2$, $σ_3$) acting along $\left<110 \right>$, $\left<1\bar{1}0 \right>$, and $\left<001\right>$, respectively, dictating the simulation cell, we determine combinations of 6 independent stresses that drive a lattice instability for the Si I$\rightarrow$Si II PT, and a semiconductor-metal electronic transition. Metallization precedes the structural PT, hence, a stressed Si I can be a metal. Surprisingly, a stress-free Si II is metastable in DFT. Notably, the PT for hydrostatic pressures is at 75.81 GPa, while under uniaxial stress it is 11.03 GPa (or 3.68 GPa mean pressure). Our key result: The Si I -> Si II PT is described by a critical value of the modified transformation work, as found with a phase-field method, and the PT criterion has only two parameters for a general applied stress. More generally, our findings are crucial for revealing novel (and more economic) material synthesis routes for new or known high-pressure phases under controlled and predictable non-hydrostatic loading and plastic deformation.

cond-mat.mtrl-sci↗

arXiv2018

SINet: A Scale-insensitive Convolutional Neural Network for Fast Vehicle Detection

Vision-based vehicle detection approaches achieve incredible success in recent years with the development of deep convolutional neural network (CNN). However, existing CNN based algorithms suffer from the problem that the convolutional features are scale-sensitive in object detection task but it is common that traffic images and videos contain vehicles with a large variance of scales. In this paper, we delve into the source of scale sensitivity, and reveal two key issues: 1) existing RoI pooling destroys the structure of small scale objects, 2) the large intra-class distance for a large variance of scales exceeds the representation capability of a single network. Based on these findings, we present a scale-insensitive convolutional neural network (SINet) for fast detecting vehicles with a large variance of scales. First, we present a context-aware RoI pooling to maintain the contextual information and original structure of small scale objects. Second, we present a multi-branch decision network to minimize the intra-class distance of features. These lightweight techniques bring zero extra time complexity but prominent detection accuracy improvement. The proposed techniques can be equipped with any deep network architectures and keep them trained end-to-end. Our SINet achieves state-of-the-art performance in terms of accuracy and speed (up to 37 FPS) on the KITTI benchmark and a new highway dataset, which contains a large variance of scales and extremely small objects.

cs.CV↗

arXiv2018

The MALATANG Survey: the L_gas-L_IR correlation on sub-kiloparsec scale in six nearby star-forming galaxies as traced by HCN J=4-3 and HCO^+ J=4-3

We present HCN J=4-3 and HCO^+ J=4-3 maps of six nearby star-forming galaxies, NGC 253, NGC 1068, IC 342, M82, M83, and NGC 6946, obtained with the James Clerk Maxwell Telescope as part of the MALATANG survey. All galaxies were mapped in the central 2 arcmin $\times$ 2 arcmin region at 14 arcsec (FWHM) resolution (corresponding to linear scales of ~ 0.2-1.0 kpc). The L_IR-L'_dense relation, where the dense gas is traced by the HCN J=4-3 and the HCO^+ J=4-3 emission, measured in our sample of spatially-resolved galaxies is found to follow the linear correlation established globally in galaxies within the scatter. We find that the luminosity ratio, L_IR/L'_dense, shows systematic variations with L_IR within individual spatially resolved galaxies, whereas the galaxy-integrated ratios vary little. A rising trend is also found between L_IR/L'_dense ratio and the warm-dust temperature gauged by the 70 μm/100 μm flux ratio. We find the luminosity ratios of IR/HCN(4-3) and IR/HCO^+(4-3), which can be taken as a proxy for the efficiency of star formation in the dense molecular gas (SFE_dense), appears to be nearly independent of the dense-gas fraction (f_dense) for our sample of galaxies. The SFE of the total molecular gas (SFE_mol) is found to increase substantially with f_dense when combining our data with that on local (ultra)luminous infrared galaxies and high-z quasars. The mean L'_HCN(4-3)/L'_HCO^+(4-3) line ratio measured for the six targeted galaxies is 0.9+/-0.6. No significant correlation is found for the L'_HCN(4-3)/L'_HCO^+(4-3) ratio with the SFR as traced by L_IR, nor with the warm-dust temperature, for the different populations of galaxies.

astro-ph.GA↗

arXiv2018

Sequential change-point detection based on nearest neighbors

We propose a new framework for the detection of change-points in online, sequential data analysis. The approach utilizes nearest neighbor information and can be applied to sequences of multivariate observations or non-Euclidean data objects, such as network data. Different stopping rules are explored, and one specific rule is recommended due to its desirable properties. An accurate analytic approximation of the average run length is derived for the recommended rule, making it an easy off-the-shelf approach for real multivariate/object sequential data monitoring applications. Simulations reveal that the new approach has better performance than likelihood-based approaches for high dimensional data. The new approach is illustrated through a real dataset in detecting global structural changes in social networks.

stat.ME↗

arXiv2018

Angora: Efficient Fuzzing by Principled Search

Fuzzing is a popular technique for finding software bugs. However, the performance of the state-of-the-art fuzzers leaves a lot to be desired. Fuzzers based on symbolic execution produce quality inputs but run slow, while fuzzers based on random mutation run fast but have difficulty producing quality inputs. We propose Angora, a new mutation-based fuzzer that outperforms the state-of-the-art fuzzers by a wide margin. The main goal of Angora is to increase branch coverage by solving path constraints without symbolic execution. To solve path constraints efficiently, we introduce several key techniques: scalable byte-level taint tracking, context-sensitive branch count, search based on gradient descent, and input length exploration. On the LAVA-M data set, Angora found almost all the injected bugs, found more bugs than any other fuzzer that we compared with, and found eight times as many bugs as the second-best fuzzer in the program who. Angora also found 103 bugs that the LAVA authors injected but could not trigger. We also tested Angora on eight popular, mature open source programs. Angora found 6, 52, 29, 40 and 48 new bugs in file, jhead, nm, objdump and size, respectively. We measured the coverage of Angora and evaluated how its key techniques contribute to its impressive performance.

cs.CR↗

arXiv2018

LSTD: A Low-Shot Transfer Detector for Object Detection

Recent advances in object detection are mainly driven by deep learning with large-scale detection benchmarks. However, the fully-annotated training set is often limited for a target detection task, which may deteriorate the performance of deep detectors. To address this challenge, we propose a novel low-shot transfer detector (LSTD) in this paper, where we leverage rich source-domain knowledge to construct an effective target-domain detector with very few training examples. The main contributions are described as follows. First, we design a flexible deep architecture of LSTD to alleviate transfer difficulties in low-shot detection. This architecture can integrate the advantages of both SSD and Faster RCNN in a unified deep framework. Second, we introduce a novel regularized transfer learning framework for low-shot detection, where the transfer knowledge (TK) and background depression (BD) regularizations are proposed to leverage object knowledge respectively from source and target domains, in order to further enhance fine-tuning with a few target images. Finally, we examine our LSTD on a number of challenging low-shot detection experiments, where LSTD outperforms other state-of-the-art approaches. The results demonstrate that LSTD is a preferable deep detector for low-shot scenarios.

cs.CV↗

arXiv2018

Use of Two-Mode Circuitry and Optimal Energy-Efficient Power Control Under Target Delay-Outage Constraints

An accurate energy efficiency analytical model based on a two-mode circuitry was recently proposed; and the model showed that the use of this circuitry can significantly improve a system's energy efficiency. In this paper, we use this analytical model to develop a new power control scheme, a scheme that is capable of allocating a minimum transmission power precisely within the delay-outage probability constraint. Precision brings substantial benefits as numerical results show that the energy efficiency using our scheme is much higher than other schemes. Results further suggest that data rate values affect energy efficiency non-uniformly, i.e., there exists a specific data rate value that achieves maximum energy efficiency.

eess.SP↗

arXiv2018

Asymptotic Distribution-Free Change-Point Detection for Multivariate and non-Euclidean Data

We consider the testing and estimation of change-points, locations where the distribution abruptly changes, in a sequence of multivariate or non-Euclidean observations. We study a nonparametric framework that utilizes similarity information among observations, which can be applied to various data types as long as an informative similarity measure on the sample space can be defined. The existing approach along this line has low power and/or biased estimates for change-points under some common scenarios. We address these problems by considering new tests based on similarity information. Simulation studies show that the new approaches exhibit substantial improvements in detecting and estimating change-points. In addition, under some mild conditions, the new test statistics are asymptotically distribution free under the null hypothesis of no change. Analytic p-value approximations to the significance of the new test statistics for the single change-point alternative and changed interval alternative are derived, making the new approaches easy off-the-shelf tools for large datasets. The new approaches are illustrated in an analysis of New York taxi data.

stat.ME↗

arXiv2018

Practical Issues of Action-conditioned Next Image Prediction

The problem of action-conditioned image prediction is to predict the expected next frame given the current camera frame the robot observes and an action selected by the robot. We provide the first comparison of two recent popular models, especially for image prediction on cars. Our major finding is that action tiling encoding is the most important factor leading to the remarkable performance of the CDNA model. We present a light-weight model by action tiling encoding which has a single-decoder feedforward architecture same as [action_video_prediction_honglak]. On a real driving dataset, the CDNA model achieves ${0.3986} \times 10^{-3}$ MSE and ${0.9846}$ Structure SIMilarity (SSIM) with a network size of about {\bfseries ${12.6}$ million} parameters. With a small network of fewer than {\bfseries ${1}$ million} parameters, our new model achieves a comparable performance to CDNA at ${0.3613} \times 10^{-3}$ MSE and ${0.9633}$ SSIM. Our model requires less memory, is more computationally efficient and is advantageous to be used inside self-driving vehicles.

cs.CV↗

arXiv2017

SFCN-OPI: Detection and Fine-grained Classification of Nuclei Using Sibling FCN with Objectness Prior Interaction

Cell nuclei detection and fine-grained classification have been fundamental yet challenging problems in histopathology image analysis. Due to the nuclei tiny size, significant inter-/intra-class variances, as well as the inferior image quality, previous automated methods would easily suffer from limited accuracy and robustness. In the meanwhile, existing approaches usually deal with these two tasks independently, which would neglect the close relatedness of them. In this paper, we present a novel method of sibling fully convolutional network with prior objectness interaction (called SFCN-OPI) to tackle the two tasks simultaneously and interactively using a unified end-to-end framework. Specifically, the sibling FCN branches share features in earlier layers while holding respective higher layers for specific tasks. More importantly, the detection branch outputs the objectness prior which dynamically interacts with the fine-grained classification sibling branch during the training and testing processes. With this mechanism, the fine-grained classification successfully focuses on regions with high confidence of nuclei existence and outputs the conditional probability, which in turn benefits the detection through back propagation. Extensive experiments on colon cancer histology images have validated the effectiveness of our proposed SFCN-OPI and our method has outperformed the state-of-the-art methods by a large margin.

cs.CV↗

arXiv2017

Optimizing relinearization in circuits for homomorphic encryption

Fully homomorphic encryption (FHE) allows an untrusted party to evaluate arithmetic cir- cuits, i.e., perform additions and multiplications on encrypted data, without having the decryp- tion key. One of the most efficient class of FHE schemes include BGV and FV schemes, which are based on the hardness of the RLWE problem. They share some common features: ciphertext sizes grow after each homomorphic multiplication; multiplication is much more costly than addition, and the cost of homomorphic multiplication scales linearly with the input ciphertext sizes. Furthermore, there is a special relinearization operation that reduce the size of a ciphertext, and the cost of relinearization is on the same order of magnitude as homomorpic multiplication. This motivates us to define a discrete optimization problem, which is to decide where (and how much) in a given circuit to relinearize, in order to minimize the total computational cost. In this paper, we formally define the relinearize problem. We prove that the problem is NP-hard. In addition, in the special case where each vertex has at most one outgoing edge, we give a polynomial-time algorithm.

cs.DS↗

arXiv2017

Minimal Twin Surfaces

We report some minimal surfaces that can be seen as copies of a triply periodic minimal surface (TPMS) related by reflections in parallel mirrors. We call them minimal twin surfaces for the resemblance with twin crystal. Brakke's Surface Evolver is employed to construct twinnings of various classical TPMS, including Schwarz' Primitive (P) and Diamond (D) surfaces, their rhombohedral deformations (rPD), and Schoen's Gyroid (G) surface. Our numerical results provide strong evidences for the mathematical existence of D twins and G twins, which are recently observed in experiment by material scientists. For rPD twins, we develop a good understanding, by noticing examples previously constructed by Traizet (2008) and Fujimori and Weber (2009). Our knowledge on G twins is, by contrast, very limited. Nevertheless, our experiments lead to new cubic polyhedral models for the D and G surfaces, inspired by which we speculate new TPMS deformations in the framework of Traizet.

math.DG↗