Understanding how the brain processes information using neural circuits remains a challenging task. For example, although neurons in the mammalian visual cortex are known to be specialized for processing particular visual features, such as motion, color or depth, little progress has been made on elucidating the underlying circuit diagrams. To address this problem, we have developed a system with a Java-based user interface that can apply optimization strategies to rapidly test competing models against large amounts of neurophysiological data, parallelizing the computational load across a cluster of workstations, to uncover plausible circuit models. In order to demonstrate the effectiveness of this system, we are applying it to understand the function of motion-sensitive neurons in the primate visual cortex. These neurons are direction selective (DS), meaning that they respond well to a visual stimulus moving in one direction but not in others. Previous studies demonstrated that this response also changes with stimulus speed: as the visual world moves faster, the neurons appear to operate on a shorter timescale. Current mathematical models for DS neurons, which are based on filters that change position linearly with time, predict the response to some stimuli but do not account for the ability of the neuronal response to adapt to changes in speed. We hypothesize that a model using a filter that changes position nonlinearly with time will better explain the experimental data. To discover this filter, we are building several candidate models and will apply our optimization system to test their performance against our own experimental data. We will test these nonlinear filter models against an alternative that uses multiple linear filters, each tuned for a different velocity. If our approach can improve upon current models for DS neurons, it may offer neuroscientists a powerful tool for revealing neural circuits that underlie visual perception.

Language models characterize the probability of a sequence of words for use in a variety of natural language processing applications, including automatic speech recognition. Web text is a key source of data for developing the language models used in such systems because of its sheer volume, and the fact that much of it is freely available to the public. However, web text varies broadly in genre, depending on purpose, audience, formality, and other characteristics. For example, the difference between the formal exposition of a Wikipedia article and the informal banter between two Twitter users. When using web text to train language models for conversational speech recognition in particular, it is important to find text that resembles that speaking style. When training a language model using various web text sources, we use a measure of worth called perplexity. By selecting web text that matches the conversational style of speech, the language model perplexity improves, which generally leads to higher accuracy speech recognition. I work on selecting text taken from Twitter. This involves research on what kinds of Twitter searches will result in tweets containing useful text, as well as how to identify Twitter users who post useful text.

As the NFL was going into a lockout in 2011, linebacker Ray Lewis made the controversial statement that crime rates would rise if the NFL season was lost. Mr. Lewis said, "There's too many people that live through us," and that without football teams to follow, those people would turn to crime. At the time, there were many arguments provided on both sides of the issue, but no empirical research was done. Studies have shown that professional athletes commit more crimes during a lockout; and that children with access to sports programs are less likely to grow up committing crimes; but no one has ever tested the claim that a professional sports team has an effect on the crime rate in the area in which it is based. My experience living in a Seahawks-enamored Seattle in 2013 has motivated me to take Mr. Lewis's assertion one step further and investigate whether the success of a pro sports team can raise or lower the crime rate in its city. I look at more than just the relationship between winning percentage and change in crime rate (note that I cannot compare nominal crime rates, due to the varying base levels of crime in metro areas as diverse as New York and Denver). Other measures I investigate are margin of victory or closeness of playoff race, as perhaps a more dramatic season gives the fans more to live vicariously through. Despite some relationship between winning percentage and playoff success of a team and the city's crime rate, I can't show a significant enough correlation to conclusively say that the NFL influences the rate of crime. While this doesn't have any direct implications on the NFL or on policy makers, I believe that my results are enough to encourage further study of the issue.

As mobile touch screen devices become increasingly ubiquitous, they are also becoming more accessible to the visually impaired. Someone who is blind traditionally enters text on his or her phone by using a screen reader and the default keyboard layout. However, this method is slow and it is difficult to correct mistakes as one is typing. We developed Perkinput, a nonvisual text entry method that uses the 6-bit Braille character encoding. In Perkinput, instead of relying on the device’s voice over system to announce each letter that is pressed, this input is based solely on the Braille keyboard. The user simultaneously presses down with his or her fingers that correspond to the dots for the character in Braille he or she wanted to enter. Once entered, the text can be sent in a text message, an email, or simply copied to the clipboard for later use in other applications. Results have shown that the Perkinput method is more than twice as fast as using the default keyboard layout for the blind. I have implemented this method into an iOS application that is currently deployed on the iTunes Store. I have instrumented this application in order to gain insight into how people are using the program outside of a lab setting. This will help us improve Perkinput and make it more accessible to the blind community. Perkinput aims to make text entry easier and efficient for people who cannot look at their screen.

North and South Korea have grown increasingly distant culturally, economically, and ideologically since the division of the Korean peninsula in 1948. Although the 1953 armistice agreement ended the Korean War, North-South tensions have persisted. Yet, North Korea often takes on a dual role in the South Korean national imaginary as not just a threat, but also a lost family member. This project asks why the desire for unification, which might have made sense after the Korean war but has become more difficult to achieve, continues to resonate in South Korea such that there is a fixation on unification as the one true solution to North-South tensions. My hypothesis posits an answer in two parts. First, unification is fixated on because national division represents the continuation of a historical pattern demonstrated by Korea’s internal response to the fall of the Ming dynasty, the colonization of Korea by Japan, and the Korean war. In each case, a traumatic event triggered an anxiety over political legitimacy. I examine how the internal response to each crisis was to remove any member of the nation, construed as family, who threatened the legitimacy of the state, to suggest that unification as a continuation of this pattern symbolizes the reconciliation of this divided family. Second, unification continues to resonate because it perpetuates the problem it attempts to resolve -- that is, unification itself reminds of this historical process wherein North Korea continues to be both family and enemy. To this end, I examine recent South Korean television dramas related to unification: first, to observe how the internal response to threats to Korean legitimacy exhibited by Korea in the historical pattern I have outlined constitutes a repeated narrative trope, and second, to analyze the tension between portrayals of North Koreans as both family and enemy in these dramas.

A species distribution model can be used to assess the effects of climate change, the impact of invasive species, and the effects of resource management decisions, however, obtaining data for a remote and rugged areas like National Parks often takes considerable resources. Luckily, our study site, Mt. Rainier National Park, has around 1.5 million visitors every year, many of which upload images of the landscape to photo sharing websites like Flickr. Each photo has a geolocation tag and a timestamp of when the photo was taken, revealing a species’ location and phenology at a given time and place. Using this data from photo sharing websites, we are developing models to predict the spatial distributions of five montane wildflowers that are located on Mt. Rainier: Arctic lupine, magenta paintbrush, American bistort, western pasqueflower, and Sitka valerian. Species distribution models are frequently used to understand the relation between environmental variables and a species range. Using environmental data from Mt. Rainier, we will be able to predict the current habitat of these five species and make predictions of how the species range will change under various climate change scenarios. Additionally, due the relative ease of accumulating data, we plan to adjust our geographic focus to other national parks where species distribution predictions could improve resource management and conservation decisions.

In the classical secretary problem, an employer tries to hire the best secretary from a line of candidates. The employer considers each candidate in sequence and makes an irreversible decision to accept or reject, being able to make only one hire. A well-known result gives the optimal strategy for this problem with a probability of success of 1/e. We extend this problem by considering a setting in which there are multiple employers, which could be universities, for example, competing for the best faculty candidates. We derive the optimal strategy for this setting and demonstrate that it meets expectations because lower ranked universities tend to make earlier offers. We employ theoretical analysis using game theory and algorithms. The broader implications of this research are greater understanding of optimal stopping problems as well as better information on how employers should hire. Eric's duties include design and analysis of algorithms for solving the problem.

Physical limitations have forced computer hardware designers into providing multiple processors rather than faster ones. To take advantage of this, software engineers must structure their programs in a parallel fashion. Separately, programming languages with commands similar to those of hardware, known as "systems languages", have long been criticized for their lack of safety. This means that programs written in these languages are easily incorrect and difficult to reason about. Rust, a new systems language from Mozilla, seeks to address both parallelism and safety by using modern language concepts. To those ends, Rust’s concurrency primitive is the task, which can be thought of as a lightweight thread that does not share memory with any other task. Message passing is used between these tasks in order to coordinate their work. While this primitive is sophisticated and suited to parallelism, the development community has expressed a desire for an Actor library. At a high level, the Actor model is a concurrent object-oriented programming model, where each Actor maintains state and takes actions based on the messages they receive from other Actors. In this investigation, the suitability of the Rust language for an Actor library is evaluated by writing and refining an implementation. It is expected that the implementation will be backed by the task primitive and its associated structures. Various criteria will be evaluated throughout development, including overall throughput, library overhead, and ease of use. The success of the Actor model within the Rust language would continue its goal of bringing straightforward parallelism and safety to the systems language space.

Exoplanets are planets outside our solar system. The discovery and characterization of exoplanet systems informs us of the types of worlds found in nature and whether they might support life. There have been an unprecedented number of exoplanet detections in recent years spurred in large part by the advanced observing capabilities of NASA’s Kepler Space Telescope. The Kepler Spacecraft continuously monitors a large, fixed set of stars and collects time series photometric data that are unparalleled in both quantity and precision. Plots of stellar brightness against time called lightcurves are the canonical representation of Kepler data. The goal of our research is to detect exoplanets from the characteristic features they produce in these lightcurves known as transits. A transit occurs when an orbiting planet partially eclipses its stellar host, reducing the brightness observed by the telescope and recorded in the lightcurve. To detect transits, we use the Quasi-Periodic Automated Transit Search Algorithm (QATS). As an automated tool, QATS is instrumental in reducing the effective size of the Kepler dataset. However, QATS is sensitive to variability of the target star and to systematic artifacts of the telescope just as it is to transits. Thus optimal detection efficiency requires removal of all non-transit signals prior to applying the algorithm. Determining the best way to do this is the present focus of our work because it bounds our ability to detect planets whose size, composition, and orbital parameters are conducive to life.

It is often forgotten students are learning to build a better future. As a group of students, we are looking to create a form of innovative technology that provides a learning experience through two of the most important learning methods: visual and kinesthetic. To provide hands on learning experience, we have created the Dynamic Mathematical Display (DMD). The DMD is a device with pins placed within a three by three grid. The pins are 10 cm long, 3.5 cm wide, and each controlled by individual motors and slide potentiometers. The dynamic movement of each pin is measured by the potentiometer's linear resistance, which is monitored electronically by a computer program. That program will engage the pin to move to a given point projected from 3D graph. The main purpose of the DMD project is to create a morphing table that will allow a user to visualize complicated 3D graphs with all their components such as depths and peaks. These components will be represented by red, green, and blue light emitting diodes (RGB LEDs), which are attached to each pin. The LED's are programmed to provide color based on the physical location of the pin at the time according to the projected graph. Both instructors and students will have the option to manipulate the brightness and color of the LEDs. While the pins provide kinesthetic information to a graph, the LEDs provide additional visual information similar to a topographic map, enhancing the learning experience of a student by presenting more realistic and information-rich representations of physical models. 

The surface of Mars is dominated by wind-blown processes and sediment deposits. Dune fields are especially ubiquitous on the surface and record important information about atmospheric circulation patterns, chemical and mechanical weathering, and sediment deposition. Therefore, dune fields are crucial in an understanding of Martian properties such as surface processes, past and present climates, and geologic history. This research investigates the thermophysical properties of sediments composing equatorial dune fields using data from the Thermal Emission Imaging System (THEMIS) on board the Mars Odyssey spacecraft. These data are accessed through JMARS, a geospatial information system developed by Arizona State University. The data are then used with other tools, including Hugh Kieffer’s KRC model, to derive the apparent thermal inertia of the fields. These thermal inertia estimates are the primary way of studying the grains for this research. A material’s temperature response during a diurnal, or daily, cycle can be used to approximate thermal inertia and subsequently be converted to an estimated sediment size. A variety of sizes is expected to be found within the dune fields. Grain size is an important characteristic that can indicate sediment source (local or regional), wind speed, and weathering history, ultimately leading to a better understanding of the surface and climate of Mars. Future research will focus on locating sediment sources and defining textural maturity for Martian sediments, which includes grain size and rounding. It will also focus on identifying inactive dunes, which may preserve historical climate indicators.

The frequency of paternal mitochondrial inheritance in humans has yet to be established. However, previous studies have indicated that such inheritance either never occurs or is very rare. Whole-genome sequences (WGS) of human trios provide a new technology to test the hypothesis that paternal mitochondrial inheritance occurs, and if so, to estimate the frequency. Population genetics and genetic diagnosis may benefit from accurate estimates of paternal mitochondrial inheritance. We test inheritance in several hundred trios by comparing the mitochondrial genomes (mtDNA) in mother-father-child trios. Paternal inheritance would be deduced if the father and child mtDNA matched, but the mother and child mtDNA did not. Although paternal inheritance is unlikely to be seen with our limited dataset, we demonstrate analysis of WGS to be effective for exploring mitochondrial inheritance in humans.

The HOPE Study is a randomized control trial conducted by investigators from the Kenya Research Program at the University of Washington Department of Global Health, Kenyatta National Hospital, and the Kenyan Ministry of Health. The clinical trial investigates the impact of home-based HIV testing and education for partners of pregnant mothers in Kisumu, Kenya. Couples are randomized to either home-based partner education and HIV testing (HOPE) as part of routine pregnancy services or to standard antenatal care (ANC). The intervention provides education and HIV testing to men in stable relationships with pregnant HIV-infected and HIV-uninfected women in order to improve overall health of women and infants, reduce risk for vertical and horizontal transmission, and increase identification of men living with HIV and link them to care. Couples will be followed up for uptake of HIV testing, facility delivery, exclusive breastfeeding and postpartum contraceptive use, as well as linkage to HIV care. Cost-effectiveness of HOPE will be evaluated in order to inform future scale up of the intervention in Western Kenya, a region with high HIV-1 incidence during the pregnant/postpartum period and high seroprevalence among men (~10%). A custom Open Data Kit (ODK) Tables application developed at the University of Washington Computer Science and Engineering Department serves as the entry-point for data collection, allowing users to collect client health information, send data to a server, and view aggregate data on their Android device. The HOPE Study Tables application allows the nurses and community health workers to screen patients for eligibility, follow up with the study participants and collect geo point data for home visits. The HOPE Study demonstrates the capabilities of ODK Tables as a data browser and manager, highlighting the importance of developing customizable data-driven applications for disconnected environments.

The current standard eyes-free speech dictation mechanism available on mobile phones is Siri's Eyes-Free mode. There are only three functions available: reviewing the current message, redoing the entire message, and sending the message. This makes speech-input of a long text-message on a mobile device very tedious. I created an app that automatically detects potential errors by aligning alternative dictations of the message and allows users to swipe through different portions of the message. I did a study on 6 users to dictate and edit sentences using Siri's Eyes-Free mode and the app that I created, and I did another study on 10 users to dictate and edit paragraphs using Siri and my app. I compare the resultsof Siri and my app in the two studies. I found that Siri's Eyes-Free mode was generally acceptable for short messages, but that my app increased efficiency for sending paragraphs. My app decreased the amount of time users spent editing paragraphs and also decreased the average word error rate.