Announcing the 2013-2014 Facebook Graduate Fellows
Thank you to many excellent students who applied to the Facebook Graduate Fellowship program. After careful consideration by our engineers, researchers, and leadership, we are excited to announce the 2013-2014 Facebook Graduate Fellowship winners and finalists.
Our fellowships support emerging research leaders who demonstrate potential to advance our mission of making the world more open and connected. Fellowships cover tuition and fees and provide a $30,000 stipend in addition to conference travel and other benefits.
Meet the winners and finalists below and check out an overview of their research. We congratulate everyone for their hard work and compelling ideas.
Facebook Graduate Fellows
Area of Focus: Distributed Systems and Databases
Sameer is a fourth-year Ph.D. student in the AMPLab at Berkeley working on large-scale approximate query processing frameworks. His research interests lie at the intersection of distributed systems, databases and machine learning. Recently, he has worked in designing dynamic parallel query optimization frameworks, building snapshot managers and proactive file replication schemes for distributed file systems, and exploring the usefulness of stateless packet classification protocols in datacenters.
Modern data analytics applications typically involve processing massive amounts of data to take quick, near real-time decisions. While the quantity of data makes arbitrary interactive queries prohibitive, Sameer observed that many analytic queries can tolerate some levels of inaccuracy. To this end, his research primarily focusses on designing large-scale interactive query processing frameworks that enable an explicit trade-off between accuracy and query execution times. Currently, his work is centered around designing BlinkDB, an open-sourced, massively parallel database that enables users to specify either an error bound (and a confidence interval) on the query result, or a time limit on the query execution-- depending on which BlinkDB either optimizes for the response time or maximizes the result accuracy.
Area of Focus: Machine Learning
Krishna is a third-year Ph.D. student at Geaorgia Tech's College of Computing where he obtained his Masters in 2010. His research focuses on several aspects of machine learning and statistics. It primarily involves formulating and solving problems which enable computers to analyze and learn from the vast amounts of real world data.
The main aim of his current research is to design and analyze flexible statistical procedures based on matrix and functional models that are suitable for analyzing modern large-scale high-dimensional data. It involves both understanding the statistical properties of the developed procedures as well as developing efficient computational algorithms. The developed methods are applicable for carrying out various statistical inference tasks successfully in several non-standard situations that often arise in practice.
Area of Focus: Natural Language Processing
Jackie is a third-year Ph.D. student at the University of Toronto, and has previously worked on other topics in natural language generation including opinion summarization and the interaction between word order and information structure.
The sheer amount of text available electronically is both a challengeand an opportunity for users who want to access some specificinformation amidst the threat of information overload. The goal of Jackie's research is to help users manage this information byproviding automatic summaries of some source text. In particular, heworks on semantic representations of text that can be used to makeinferences about what is salient or important in a domain for use incomplex tasks such as summarization.
Area of focus: Human-Computer Interaction
Lydia has been doing crowdsourcing since 2008 as an undergraduate and Master's student at the Massachusetts Institute of Technology where she did a year long internship in Beijing working closely with designers and researchers. Lydia is currently a fourth-year Ph.D. student at the University of Washington studying under James Landay and Dan Weld.
How can we use the crowd to organize data? Lydia is writing algorithms to coordinate crowd labor. When you get a group of people working together, sometimes you get great work, and sometimes you get a lot of noise. Lydia's work focuses on finding primitives used in crowd algorithms that reliably get strong signal instead of noise.
Area of Focus: Human-Computer Interaction
Justin is a fourth-year Ph.D. student in the School of Computer Science at Carnegie Mellon, where his research contributes to a number of fields, including ubiquitous computing, human-computer interaction, and computational social science.
Justin wants to build systems that can make sense of the complex and ever changing landscape of big cities. By teaching computers to understand and interpret the urban environment just as well as humans, Justin believes we can unleash a whole new wave of social and mobile applications that improve the way we live, work, and play in cities.
Area of Focus: Human-Computer Interaction
Nicki is a fourth-year Ph.D. student in the Department of Computer Science and Engineering at the University of Washington. Her research focuses on how technology can improve the lives of underserved populations in low-income regions.
Nicki's thesis will explore the potential for computer vision and machine-learning algorithms running on commercially available smartphones to improve data collection and disease diagnosis in remote areas. In addition to the technical challenges posed by designing these algorithms, Nicki is also committed to studying the human-computer interaction issues that arise through real deployments with rural users in remote areas of the world. She therefore spends substantial amounts of time working with users in the field to ensure that the technology is usable and appropriate under the constraints experienced in low-resource settings.
Area of Focus: Distributed Systems
Qi is a Ph.D. student in the Systems Lab at Cornell University's Department of Computer Science. Qi's research interest lies in the area ofdistributed systems, and the projects he has worked also intersect with networking and storage in the context of cloud computing.
Currently, Qi is tackling emerging adaptivity problems seenin today's massive-scale distributed systems, which often run indiverse environments. Such environments can arise as a consequence ofdifferent management domains, various network layouts, or from diverseapplication scenarios each bringing its own load patterns. When amonolithic design gets exposed to multiple distinct environments, theinability to adapt can put the whole system under stress and disruptapplications on top. Qi's goal is to bring environmental awarenessinto the design of distributed system, so that systems can adaptthemselves without disrupting services.
Area of Focus: Computer Vision
Aditya is a second-year Ph.D. student in computer science at the Massachusetts Institute of Technology. He completed his M.S. at Stanford University in 2011 and B.S. at the California Institute of Technology in 2009. His current research focus is on applying computer vision and machine learning techniques to human memory. Recent research shows that the extent to which an image is remembered is largely a property of the image and not of the individual. This means that despite varied experiences, individuals tend to remember and forget the same images.
From the increasing number of photos being viewed and shared on the internet, what do people remember? Can we predict what makes an image memorable or forgettable? What can be changed in an image to increase or decrease its memorability? Aditya's research aims to answer these questions. He is currently working on devising algorithms to allow the automatic modification of image memorability. This would be akin to having a new filter on Instagram called "Memorable" that makes certain modifications to an image or a face to make it more or less memorable. This technology could have significant impact on the way we design images for various applications e.g. advertisers could make their advertising more memorable and thereby more effective, or teachers could modify textbook diagrams to allow them to be easily remembered by students. For more information, visit his website: http://mit.edu/khosla
Area of Focus: Distributed Systems
Crystal is a first-year Ph.D. student at Stanford University. Her research aims to bring techniques from compilers and programming languages to the design and implementation of language and compiler tools that help programmers manipulate and analyze problems with large data sets.
Performing computations on large data sets often require complex, specially designed computer systems that make tradeoffs between features like multiple hardware threads, vector‐width computations, and faster, larger memory systems – features that can make these architectures very difficult to program. Domain-specific languages are a technique for addressing this difficulty: they allow language designers to separate the high‐level expression of an algorithm from systems‐level concerns and shift the work of efficiently parallelizing the algorithm onto the compiler. Crystal is working on developing tools that enable language designers to build DSLs and demonstrate that they are a serious practical solution to enable higher level, non-systems programmers to leverage complex architectures and perform computations on their data.
Area of Focus: Architecture
Chao believes that greening computing is a win-win proposition. His research intends to demonstrate how IT design and operation can be sustainable and social responsible. Inspired by the Open Compute Project of Facebook, Chao aims to design the greenest computing infrastructures with the lowest overhead. Going beyond conventional approaches, he looks at novel system design that integrates emerging hardware and system techniques with promising green energy solutions. His research enables datacenters to integrate renewable energy more intelligently and thus, paves the way for the next-generation real green computing platform.
Chao received the IEEE HPCA 2011 Best Paper Award for devising a solar energy driven multi-core architecture power management scheme. He is currently a fourth-year Ph.D. student in Computer Engineering at the University of Florida. Prior to coming to UF, Chao earned his BS in Electrical Engineering from Zhejiang University, one of the most prestigious universities in China and is an alumnus of Chu Kochen Honors College, an undergraduate program at Zhejiang University.
Area of Focus: Theory
Currently a fourth-year Ph.D. student at Carnegie Mellon University, Julian obtained his undergraduate degree from UC Berkeley. He is doing research in parallel algorithms for shared memory multicore machines. Due to the need to quickly process large data, Julian is particularly interested in developing large-scale parallel algorithms. He is developing Ligra, a lightweight interface for programming large-scale graph algorithms for shared memory. He is also doing work on parallel string algorithms, which are important for quickly processing large data.
Additionally, Julian is interested in studying techniques that simplify shared memory programming. To this end, he has developed methods for writing deterministic parallel programs, which simplify programming and debugging. He is also developing the Problem Based Benchmark Suite, which allows for head-to-head comparison of parallel languages and architectures for the same problem.
Area of Focus: Data Mining
Bob obtained a Diplom degree in computer science from the Technical University of Munich in his native Germany in 2007 and a Master's degree in computer science from McGill University in 2010. He is currently a third-year Ph.D. student in the InfoLab at Stanford University, advised by Jure Leskovec.
Not only has the Internet revolutionized human life, it has also revolutionized the study of human life. Whereas people's pre-Web, offline behaviors went largely undocumented, nearly every onlineaction is recorded in log files. This way, human behavior can now inprinciple be analyzed in unprecedented detail. In his research, Bob has been working at the intersection of data mining, machine learning,and natural language processing to convert raw log data intomeaningful insights on a number of human behaviors, ranging fromnavigation in complex networks to Wikipedia editing to food intake.
Facebook Graduate Fellowship Finalists
Alex Beutel, Carnegie Mellon University
Yingyi Bu, University of California, Irvine
Justin Cheng, Stanford University
Raymond Cheng, University of Washington
Arlen Cox, University of Colorado at Boulder
Christina Delimitrou, Stanford University
Golnaz Ghasemiesfeh, Stony Brook University
Rebecca Balebako, Carnegie Mellon University
Kevin Jamieson, University of Wisconsin - Madison
Ilan Kadar, Ben-Gurion University of the Negev
Sanjay Kairam, Stanford University
Georgios Kontaxis, Columbia University
Paraschos Koutris, University of Washington
Walid Krichene, University of California, Berkeley
Chinmay Kulkarni, Stanford University
Katyaini Himabindu Lakkaraju, Stanford University
Reut Levi, Tel-Aviv University
Wang Ling, Carnegie Mellon University and Instituto Superior Tecnico
Wei Lu, University of British Columbia
Justin Meza, Carnegie Mellon University
Arjun Mukherjee, University of Illinois at Urbana-Champaign
Arjun Narayan, University of Pennsylvania
Genevieve Patterson, Brown University
Mohammad Rastegari, University of Maryland at College Park
Rifat Shahriyar, Australian National University
Ran Shorrer, Harvard University
Chi Wang, University of Illinois at Urbana-Champaign