Hantian Zhang

Generative adversarial networks recover features in astrophysical images of galaxies beyond the deconvolution limit

Observations of astrophysical objects such as galaxies are limited by various sources of random and systematic noise from the sky background, the optical system of the telescope and the detector used to record the data. Conventional deconvolution techniques are limited in their ability to recover features in imaging data by the Shannon-Nyquist sampling theorem. Here we train a generative adversarial network (GAN) on a sample of

Generative adversarial networks as a tool to recover structural information from cryo-electron microscopy data

4,550

MLbench: benchmarking machine learning services against human experts

images of nearby galaxies at

psfgan: a generative adversarial network system for separating quasar point sources and host galaxy light

0.01<z<0.02

Scalable inference of decision tree ensembles: Flexible design for CPU-FPGA platforms

from the Sloan Digital Sky Survey and conduct

The ZipML Framework for Training Models with End-to-End Low Precision

10\times

ZipML: Training Linear Models with End-to-End Low Precision, and a Little Bit of Deep Learning

cross validation to evaluate the results. We present a method using a GAN trained on galaxy images that can recover features from artificially degraded images with worse seeing and higher noise than the original with a performance which far exceeds simple deconvolution. The ability to better recover detailed features such as galaxy morphology from low-signal-to-noise and low angular resolution imaging data significantly increases our ability to study existing data sets of astrophysical objects as well as future observations with observatories such as the Large Synoptic Sky Telescope (LSST) and the Hubble and James Webb space telescopes.

ZipML: An End-to-end Bitwise Framework for Dense Generalized Linear Models.

Cryo-electron microscopy (cryo-EM) is a powerful structural biology technique capable of determining atomic-resolution structures of biological macromolecules. Despite this ability, the low signal-to-noise ratio of cryo-EM data continues to remain a hurdle for assessing raw cryo-EM micrographs and subsequent image analysis. To help address this problem, we have performed proof-of-principle studies with generative adversarial networks, a form of artificial intelligence, to denoise individual particles. This approach effectively recovers global structural information for both synthetic and real cryo-EM data, facilitating per-particle assessment from noisy raw images. Our results suggest that generative adversarial networks may be able to provide an approach to denoise raw cryo-EM images to facilitate particle selection and raw particle interpretation for single particle and tomography cryo-EM data.

The ZipML Framework for Training Models with End-to-End Low Precision: The Cans, the Cannots, and a Little Bit of Deep Learning.

Modern machine learning services and systems are complicated data systems --- the process of designing such systems is an art of compromising between functionality, performance, and quality. Providing different levels of system supports for different functionalities, such as automatic feature engineering, model selection and ensemble, and hyperparameter tuning, could improve the quality, but also introduce additional cost and system complexity. In this paper, we try to facilitate the process of asking the following type of questions: How much will the users lose if we remove the support of functionality x from a machine learning service? Answering this type of questions using existing datasets, such as the UCI datasets, is challenging. The main contribution of this work is a novel dataset, MLBench, harvested from Kaggle competitions. Unlike existing datasets, MLBench contains not only the raw features for a machine learning task, but also those used by the winning teams of Kaggle competitions. The winning features serve as a baseline of best human effort that enables multiple ways to measure the quality of machine learning services that cannot be supported by existing datasets, such as relative ranking on Kaggle and relative accuracy compared with best-effort systems. We then conduct an empirical study using MLBench to understand example machine learning services from Amazon and Microsoft Azure, and showcase how MLBench enables a comparative study revealing the strength and weakness of these existing machine learning services quantitatively and systematically. The full version of this paper can be found at arxiv.org/abs/1707.09562

How good are machine learning clouds for binary classification with good features?: extended abstract

The study of unobscured active galactic nuclei (AGN) and quasars depends on the reliable decomposition of the light from the AGN point source and the extended host galaxy light. The problem is typically approached using parametric fitting routines using separate models for the host galaxy and the point spread function (PSF). We present a new approach using a Generative Adversarial Network (GAN) trained on galaxy images. We test the method using Sloan Digital Sky Survey (SDSS) r-band images with artificial AGN point sources added which are then removed using the GAN and with parametric methods using GALFIT. When the AGN point source PS is more than twice as bright as the host galaxy, we find that our method, PSFGAN, can recover PS and host galaxy magnitudes with smaller systematic error and a lower average scatter (