Search

In medical imaging, access to data is commonly limited due to patient privacy restrictions and the issue that it can be difficult to acquire enough data in the case of rare diseases.[1] The purpose of this investigation was to develop a reusable open-source synthetic image generation pipeline, the GAN Image Synthesis Tool (GIST), that is easy to use as well as easy to deploy. The pipeline helps to improve and standardize AI algorithms in the digital health space by generating high quality synthetic image data that is not linked to specific patients. Its image generation capabilities include the ability to generate imaging of pathologies or injuries with low incidence rates. This improvement of digital health AI algorithms could improve diagnostic accuracy, aid in patient care, decrease medicolegal claims, and ultimately decrease the overall cost of healthcare. The pipeline builds on existing Generative Adversarial Networks (GANs) algorithms, and preprocessing and evaluation steps were included for completeness. For this work, we focused on ensuring the pipeline supports radiography, with a focus on synthetic knee and elbow x-ray images. In designing the pipeline, we evaluated the performance of current GAN architectures, studying the performance on available x-ray data. We show that the pipeline is capable of generating high quality and clinically relevant images based on a lay person's evaluation and the Fr\'echet Inception Distance (FID) metric.

We present backpropagation clipping, a novel variant of differentially private stochastic gradient descent (DP-SGD) for privacy-preserving deep learning. Our approach clips each trainable layer's inputs (during the forward pass) and its upstream gradients (during the backward pass) to ensure bounded global sensitivity for the layer's gradient; this combination replaces the gradient clipping step in existing DP-SGD variants. Our approach is simple to implement in existing deep learning frameworks. The results of our empirical evaluation demonstrate that backpropagation clipping provides higher accuracy at lower values for the privacy parameter $\epsilon$ compared to previous work. We achieve 98.7% accuracy for MNIST with $\epsilon = 0.07$ and 74% accuracy for CIFAR-10 with $\epsilon = 3.64$., Comment: We found a bug in our implementation code that invalidates our experimental results

Many software analysis methods have come to rely on machine learning approaches. Code segmentation - the process of decomposing source code into meaningful blocks - can augment these methods by featurizing code, reducing noise, and limiting the problem space. Traditionally, code segmentation has been done using syntactic cues; current approaches do not intentionally capture logical content. We develop a novel deep learning approach to generate logical code segments regardless of the language or syntactic correctness of the code. Due to the lack of logically segmented source code, we introduce a unique data set construction technique to approximate ground truth for logically segmented code. Logical code segmentation can improve tasks such as automatically commenting code, detecting software vulnerabilities, repairing bugs, labeling code functionality, and synthesizing new code., Comment: SEKE2019 Conference Full Paper

Code search and comprehension have become more difficult in recent years due to the rapid expansion of available source code. Current tools lack a way to label arbitrary code at scale while maintaining up-to-date representations of new programming languages, libraries, and functionalities. Comprehensive labeling of source code enables users to search for documents of interest and obtain a high-level understanding of their contents. We use Stack Overflow code snippets and their tags to train a language-agnostic, deep convolutional neural network to automatically predict semantic labels for source code documents. On Stack Overflow code snippets, we demonstrate a mean area under ROC of 0.957 over a long-tailed list of 4,508 tags. We also manually validate the model outputs on a diverse set of unlabeled source code documents retrieved from Github, and we obtain a top-1 accuracy of 86.6%. This strongly indicates that the model successfully transfers its knowledge from Stack Overflow snippets to arbitrary source code documents., Comment: MASES 2018 Publication

Descriptive comments play a crucial role in the software engineering process. They decrease development time, enable better bug detection, and facilitate the reuse of previously written code. However, comments are commonly the last of a software developer's priorities and are thus either insufficient or missing entirely. Automatic source code summarization may therefore have the ability to significantly improve the software development process. We introduce a novel encoder-decoder model that summarizes source code, effectively writing a comment to describe the code's functionality. We make two primary innovations beyond current source code summarization models. First, our encoder is fully language-agnostic and requires no complex input preprocessing. Second, our decoder has an open vocabulary, enabling it to predict any word, even ones not seen in training. We demonstrate results comparable to state-of-the-art methods on a single-language data set and provide the first results on a data set consisting of multiple programming languages., Comment: ENASE 2019

Using the most comprehensive source of commercially available data on the US National Market System, we analyze all quotes and trades associated with Dow 30 stocks in 2016 from the vantage point of a single and fixed frame of reference. We find that inefficiencies created in part by the fragmentation of the equity marketplace are relatively common and persist for longer than what physical constraints may suggest. Information feeds reported different prices for the same equity more than 120 million times, with almost 64 million dislocation segments featuring meaningfully longer duration and higher magnitude. During this period, roughly 22% of all trades occurred while the SIP and aggregated direct feeds were dislocated. The current market configuration resulted in a realized opportunity cost totaling over $160 million when compared with a single feed, single exchange alternative---a conservative estimate that does not take into account intra-day offsetting events., Comment: 30 pages, 9 tables, 10 figures

Both the scientific community and the popular press have paid much attention to the speed of the Securities Information Processor, the data feed consolidating all trades and quotes across the US stock market. Rather than the speed of the Securities Information Processor, or SIP, we focus here on its accuracy. Relying on Trade and Quote data, we provide various measures of SIP latency relative to high-speed data feeds between exchanges, known as direct feeds. We use first differences to highlight not only the divergence between the direct feeds and the SIP, but also the fundamental inaccuracy of the SIP. We find that as many as 60 percent or more of trades are reported out of sequence for stocks with high trade volume, therefore skewing simple measures such as returns. While not yet definitive, this analysis supports our preliminary conclusion that the underlying infrastructure of the SIP is currently unable to keep pace with the trading activity in today's stock market., Comment: 18 pages, 20 figures, 2 tables

Observations made during the New Horizons flyby provide a detailed snapshot of the current state of Pluto's atmosphere. While the lower atmosphere (at altitudes <200 km) is consistent with ground-based stellar occultations, the upper atmosphere is much colder and more compact than indicated by pre-encounter models. Molecular nitrogen (N$_2$) dominates the atmosphere (at altitudes <1800 km or so), while methane (CH$_4$), acetylene (C$_2$H$_2$), ethylene (C$_2$H$_4$), and ethane (C$_2$H$_6$) are abundant minor species, and likely feed the production of an extensive haze which encompasses Pluto. The cold upper atmosphere shuts off the anticipated enhanced-Jeans, hydrodynamic-like escape of Pluto's atmosphere to space. It is unclear whether the current state of Pluto's atmosphere is representative of its average state--over seasonal or geologic time scales., Comment: in Science 351, aad8866 (2016)

As antivirus and network intrusion detection systems have increasingly proven insufficient to detect advanced threats, large security operations centers have moved to deploy endpoint-based sensors that provide deeper visibility into low-level events across their enterprises. Unfortunately, for many organizations in government and industry, the installation, maintenance, and resource requirements of these newer solutions pose barriers to adoption and are perceived as risks to organizations' missions. To mitigate this problem we investigated the utility of agentless detection of malicious endpoint behavior, using only the standard build-in Windows audit logging facility as our signal. We found that Windows audit logs, while emitting manageable sized data streams on the endpoints, provide enough information to allow robust detection of malicious behavior. Audit logs provide an effective, low-cost alternative to deploying additional expensive agent-based breach detection systems in many government and industrial settings, and can be used to detect, in our tests, 83% percent of malware samples with a 0.1% false positive rate. They can also supplement already existing host signature-based antivirus solutions, like Kaspersky, Symantec, and McAfee, detecting, in our testing environment, 78% of malware missed by those antivirus systems.

The European Space Agency's Rosetta spacecraft, en route to a 2014 encounter with comet 67P/Churyumov-Gerasimenko, made a gravity assist swing-by of Mars on 25 February 2007, closest approach being at 01:54UT. The Alice instrument on board Rosetta, a lightweight far-ultraviolet imaging spectrograph optimized for in situ cometary spectroscopy in the 750-2000 A spectral band, was used to study the daytime Mars upper atmosphere including emissions from exospheric hydrogen and oxygen. Offset pointing, obtained five hours before closest approach, enabled us to detect and map the HI Lyman-alpha and Lyman-beta emissions from exospheric hydrogen out beyond 30,000 km from the planet's center. These data are fit with a Chamberlain exospheric model from which we derive the hydrogen density at the 200 km exobase and the H escape flux. The results are comparable to those found from the the Ultraviolet Spectrometer experiment on the Mariner 6 and 7 fly-bys of Mars in 1969. Atomic oxygen emission at 1304 A is detected at altitudes of 400 to 1000 km above the limb during limb scans shortly after closest approach. However, the derived oxygen scale height is not consistent with recent models of oxygen escape based on the production of suprathermal oxygen atoms by the dissociative recombination of O2+., Comment: 17 pages, 8 figures, accepted for publication in Icarus

The New Horizons ALICE instrument is a lightweight (4.4 kg), low-power (4.4 Watt) imaging spectrograph aboard the New Horizons mission to Pluto/Charon and the Kuiper Belt. Its primary job is to determine the relative abundances of various species in Pluto's atmosphere. ALICE will also be used to search for an atmosphere around Pluto's moon, Charon, as well as the Kuiper Belt Objects (KBOs) that New Horizons hopes to fly by after Pluto-Charon, and it will make UV surface reflectivity measurements of all of these bodies as well. The instrument incorporates an off-axis telescope feeding a Rowland-circle spectrograph with a 520-1870 angstroms spectral passband, a spectral point spread function of 3-6 angstroms FWHM, and an instantaneous spatial field-of-view that is 6 degrees long. Different input apertures that feed the telescope allow for both airglow and solar occultation observations during the mission. The focal plane detector is an imaging microchannel plate (MCP) double delay-line detector with dual solar-blind opaque photocathodes (KBr and CsI) and a focal surface that matches the instrument's 15-cm diameter Rowland-circle. In what follows, we describe the instrument in greater detail, including descriptions of its ground calibration and initial in flight performance., Comment: 24 pages, 29 figures, 2 tables; To appear in a special volume of Space Science Reviews on the New Horizons mission

The New Horizons spacecraft will achieve a wide range of measurement objectives at the Pluto system, including color and panchromatic maps, 1.25-2.50 micron spectral images for studying surface compositions, and measurements of Pluto's atmosphere (temperatures, composition, hazes, and the escape rate). Additional measurement objectives include topography, surface temperatures, and the solar wind interaction. The fulfillment of these measurement objectives will broaden our understanding of the Pluto system, such as the origin of the Pluto system, the processes operating on the surface, the volatile transport cycle, and the energetics and chemistry of the atmosphere. The mission, payload, and strawman observing sequences have been designed to acheive the NASA-specified measurement objectives and maximize the science return. The planned observations at the Pluto system will extend our knowledge of other objects formed by giant impact (such as the Earth-moon), other objects formed in the outer solar system (such as comets and other icy dwarf planets), other bodies with surfaces in vapor-pressure equilibrium (such as Triton and Mars), and other bodies with N2:CH4 atmospheres (such as Titan, Triton, and the early Earth)., Comment: 40 pages, 9 figures, 7 tables; To appear in a special volume of Space Science Reviews on the New Horizons mission

We report on spectroscopic observations of periodic comet 9P/Tempel 1 by the Alice ultraviolet spectrograph on the Rosetta spacecraft in conjunction with NASA's Deep Impact mission. Our objectives were to measure an increase in atomic and molecular emissions produced by the excavation of volatile sub-surface material. We unambiguously detected atomic oxygen emission from the quiescent coma but no enhancement at the 10% (1-sigma) level following the impact. We derive a quiescent water production rate of 9 x 10^27 molecules per second with an estimated uncertainty of 30%. Our upper limits to the volatiles produced by the impact are consistent with other estimates., Comment: 11 pages, 4 postscript figures. Accepted for publication in Icarus special issue on Deep Impact