arXiv subjects

Xiaofeng Zhang

Publications and source records attributed to Xiaofeng Zhang.

96 records · Page 6Linked to original sources

arXiv2018

Structure-Preserving Transformation: Generating Diverse and Transferable Adversarial Examples

Adversarial examples are perturbed inputs designed to fool machine learning models. Most recent works on adversarial examples for image classification focus on directly modifying pixels with minor perturbations. A common requirement in all these works is that the malicious perturbations should be small enough (measured by an L_p norm for some p) so that they are imperceptible to humans. However, small perturbations can be unnecessarily restrictive and limit the diversity of adversarial examples generated. Further, an L_p norm based distance metric ignores important structure patterns hidden in images that are important to human perception. Consequently, even the minor perturbation introduced in recent works often makes the adversarial examples less natural to humans. More importantly, they often do not transfer well and are therefore less effective when attacking black-box models especially for those protected by a defense mechanism. In this paper, we propose a structure-preserving transformation (SPT) for generating natural and diverse adversarial examples with extremely high transferability. The key idea of our approach is to allow perceptible deviation in adversarial examples while keeping structure patterns that are central to a human classifier. Empirical results on the MNIST and the fashion-MNIST datasets show that adversarial examples generated by our approach can easily bypass strong adversarial training. Further, they transfer well to other target models with no loss or little loss of successful attack rate.

cs.LG↗

arXiv2018

DADA: Deep Adversarial Data Augmentation for Extremely Low Data Regime Classification

Deep learning has revolutionized the performance of classification, but meanwhile demands sufficient labeled data for training. Given insufficient data, while many techniques have been developed to help combat overfitting, the challenge remains if one tries to train deep networks, especially in the ill-posed extremely low data regimes: only a small set of labeled data are available, and nothing -- including unlabeled data -- else. Such regimes arise from practical situations where not only data labeling but also data collection itself is expensive. We propose a deep adversarial data augmentation (DADA) technique to address the problem, in which we elaborately formulate data augmentation as a problem of training a class-conditional and supervised generative adversarial network (GAN). Specifically, a new discriminator loss is proposed to fit the goal of data augmentation, through which both real and augmented samples are enforced to contribute to and be consistent in finding the decision boundaries. Tailored training techniques are developed accordingly. To quantitatively validate its effectiveness, we first perform extensive simulations to show that DADA substantially outperforms both traditional data augmentation and a few GAN-based options. We then extend experiments to three real-world small labeled datasets where existing data augmentation and/or transfer learning strategies are either less effective or infeasible. All results endorse the superior capability of DADA in enhancing the generalization ability of deep networks trained in practical extremely low data regimes. Source code is available at https://github.com/SchafferZhang/DADA.

cs.CV↗

arXiv2018

In-flight energy calibration of the space-borne Compton polarimeter POLAR

POLAR is a compact wide-field space-borne detector for precise measurements of the linear polarisation of hard X-rays emitted by transient sources in the energy range from 50 keV to 500 keV. It consists of a 40$\times$40 array of plastic scintillator bars used as a detection material. The bars are grouped in 25 detector modules. The energy range sensitivity of POLAR is optimized to match with the prompt emission photons from the gamma-ray bursts (GRBs). Polarization measurements of the prompt emission would probe source geometries, emission mechanisms and magnetic structures in GRB jets. The instrument can also detect hard X-rays from solar flares and be used for precise measurement of their polarisation. POLAR was launched into a low Earth orbit on-board the Chinese space-lab TG-2 on September 15th, 2016. To achieve high accuracies in polarisation measurements it is essential to assure both before and during the flight a precise energy calibration. Such calibrations are performed with four low activity $^{22}$Na radioactive sources placed inside the instrument. Energy conversion factors are related to Compton edge positions from the collinear annihilation photons from the sources. This paper presents main principles of the in-flight calibration, describes studies of the method based on Monte Carlo simulations and its laboratory verification and finally provides some observation results based on the in-flight data analysis.

physics.ins-det↗

arXiv2017

Towards well-specified semi-supervised model-based classifiers via structural adaptation

Semi-supervised learning plays an important role in large-scale machine learning. Properly using additional unlabeled data (largely available nowadays) often can improve the machine learning accuracy. However, if the machine learning model is misspecified for the underlying true data distribution, the model performance could be seriously jeopardized. This issue is known as model misspecification. To address this issue, we focus on generative models and propose a criterion to detect the onset of model misspecification by measuring the performance difference between models obtained using supervised and semi-supervised learning. Then, we propose to automatically modify the generative models during model training to achieve an unbiased generative model. Rigorous experiments were carried out to evaluate the proposed method using two image classification data sets PASCAL VOC'07 and MIR Flickr. Our proposed method has been demonstrated to outperform a number of state-of-the-art semi-supervised learning approaches for the classification task.

cs.LG↗

arXiv2016

A crosstalk and non-uniformity correction method for the Compact Space-borne Compton Polarimeter POLAR

In spite of extensive observations and numerous theoretical studies in the past decades several key questions related with Gamma-Ray Bursts (GRB) emission mechanisms are still to be answered. Precise detection of the GRB polarization carried out by dedicated instruments can provide new data and be an ultimate tool to unveil their real nature. A novel space-borne Compton polarimeter POLAR onboard the Chinese space station TG2 is designed to measure linear polarization of gamma-rays arriving from GRB prompt emissions. POLAR uses plastics scintillator bars (PS) as gamma-ray detectors and multi-anode photomultipliers (MAPMTs) for readout of the scintillation light. Inherent properties of such detection systems are crosstalk and non-uniformity. The crosstalk smears recorded energy over multiple channels making both non-uniformity corrections and energy calibration more difficult. Rigorous extraction of polarization observable requires to take such effects properly into account. We studied influence of the crosstalk on energy depositions during laboratory measurements with X-ray beams. A relation between genuine and recorded energy was deduced using an introduced model of data analysis. It postulates that both the crosstalk and non-uniformities can be described with a single matrix obtained in calibrations with mono-energetic X- and gamma-rays. Necessary corrections are introduced using matrix based equations allowing for proper evaluation of the measured GRB spectra. Validity of the method was established during dedicated experimental tests. The same approach can be also applied in space utilizing POLAR internal calibration sources. The introduced model is general and with some adjustments well suitable for data analysis from other MAPMT-based instruments.

physics.ins-det↗

arXiv2014

Surface atomic diffusion processes observed at milliseconds time resolution using environmental TEM

Significant progress has been made in spatial resolution using environmental transmission electron microscopes (ETEM), which now enables atomic resolution visualization of structural transformation under variable temperature and gas environments close to materials real operational conditions. Structural transformations are observed by recording images or diffraction patterns at various time intervals using a video camera or by taking snap shots using electron pulses. While time resolution at 15 ns has been reported using pulsed electron beams, the time interval that can be recorded by this technique is currently very limited. For longer recording, however, time resolution inside ETEM has been limited by electron cameras to ~1/30 seconds for a long time. Using the recently developed direct electron detection technology, we have significantly improved the time resolution of ETEM to 2.5 ms (milliseconds) for full frame or 0.625 ms for 0.25 frames.

cond-mat.mtrl-sci↗