The Microsoft Research PhD Fellowship has supported 142 fellows since the program was established in 2008, many of whom have gone on to work within the Microsoft Research organization. Others have gone on to perform pioneering research elsewhere within the technology industry or accept faculty appointments at leading universities.
Provisions of the 2020 award
- Tuition and fees are covered for two academic years (2020–21 and 2021–22).
- A $42,000 USD stipend is provided to help with living expenses while in school for two academic years (2020-21 and 2021-22). The stipend is not expected to cover all living expenses; it can be used for expenses including, but not limited to, childcare, conference fees and travel, research equipment, meals, rent, etc.
- An invitation to interview for one salaried internship in 2020 with leading Microsoft researchers working on cutting-edge projects related to the recipient’s field of study.
- An invitation to the PhD Summit: a two-day workshop in the fall at our Redmond lab where fellows will meet with Microsoft researchers and other top students to share their research.
- Microsoft’s mission is to empower every person and every organization on the planet to achieve more. Fellows should support this mission and embrace opportunities to foster diverse and inclusive cultures within their communities.
- PhD students must be nominated by their university, and their nomination must be submitted by the office of the chair of the department.
- Students must be enrolled at a university in the United States, Canada, or Mexico.
- Proposed research must be closely related to the general research areas carried out by Microsoft Research as noted in the Research areas tab above.
- Students must be in their third year of a PhD program in the fall semester or quarter of 2019. The department chair’s office at the nominating university will need to attest that the student is considered a third year PhD student having taken into account transfers, approved leaves of absence, etc.
- A maximum of three nominations per department will be accepted; if more than two are nominated, then at least one nominee should help us increase the opportunities for students who are underrepresented in the field of computing. This includes those who self-identify as a woman, African American, Black, Hispanic, Latinx, American Indian, Alaska Native, Native Hawaiian, Pacific Islander, and/or person with a disability.
- The recipient must remain an active, full-time student in a PhD program during the two consecutive academic years of the award or forfeit the award. Fellowships are not available for extension; however, leaves of absence will be considered on a case by case basis.
- Payment of the award, as described above, will be made directly to the university and dispersed according to the university’s policies. Microsoft will have discretion as to how any remaining funds will be used if the student is no longer qualified to receive funding (e.g. if the student unenrolls from the program, graduates, or transfers to a different university).
- A recipient of the Microsoft Research PhD Fellowship may not receive another fellowship from another company or institution during the same academic period. Fellows accepting multiple fellowships may become ineligible to receive continued funding from Microsoft. Microsoft will at its sole discretion consider a joint fellowship with a government or non-profit organization.
- Nominations from the university accepted through August 15, 2019
- PhD student nominees receive a request to submit their proposal at the end of August 2019
- Proposals accepted through September 20, 2019
- Reference letters accepted through September 30, 2019
- Finalists will be notified in early November 2019
- Finalists travel to Redmond, WA for in-person interviews in late November or early December 2019
- Recipients announced by January 31, 2020
Microsoft actively seeks to foster greater levels of diversity in our workforce and in our pipeline of future researchers. We are always looking for the best and brightest talent and celebrate individuality. We invite candidates to come as they are and do what they love.
How to submit a nomination
If you are submitting a nomination on behalf of the department chair’s office at your university, the below outlines the information necessary to submit the nomination by Thursday, August 15, 2019 at 5:00 PM Pacific Time.
You will need to collect all of the below information before submitting the nomination as you will not have the ability to save the details and return to the online form. A maximum of three nominations per department will be accepted; if more than two are nominated, then at least one nominee should help us increase the opportunities for students who are underrepresented in the field of computing. This includes those who self-identify as a woman, African American, Black, Hispanic, Latinx, American Indian, Alaska Native, Native Hawaiian, Pacific Islander, and/or person with a disability.
- Nominations must include:
- Your name, email, job title, country, university, and department as the person submitting the nomination on behalf of the department chair’s office at your university
- Nominee’s name and email
- Nominee’s primary and secondary areas of research (click on the Research areas tab at the top of the page for a full list)
- Attestation that as of the fall semester or quarter of 2019, the university considers the nominee a third year PhD student (having taken into account transfers, approved leaves of absence, etc.)
How to submit a proposal
If you were nominated by your department chair’s office at your university, then you will receive an email from Microsoft Research Fellowship Program (email@example.com) by the end of August 2019 which includes a private link to submit your proposal.
If you are a nominee, the below outlines the information necessary to submit your proposal by Friday, September 20, 2019 at 5:00 PM Pacific Time. You will need to collect all of the below information before submitting your proposal as you will not have the ability to save the details and return to the online form.
- Proposals must include:
- Curriculum vitae
- Thesis proposal or research statement (short and concise is recommended—no more than five pages including references with font no smaller than 10-point font)
- One-page summary of the above thesis proposal or research statement
- Your name, email, country, university, and department
- Primary and secondary areas of research (click on the Research areas tab at the top of the page for a full list)
- Thesis proposal or research statement title
- Month and year entered the PhD program and expected graduation date (nominee must currently be in their third year of the PhD program and vetted by the university)
- Whether or not you are a doctoral student who is underrepresented in the field of computing which include those who self-identify as a woman, African American, Black, Hispanic, Latinx, American Indian, Alaska Native, Native Hawaiian, Pacific Islander, and/or person with a disability – if more than two students are nominated, then at least one should help us increase the opportunities for students who are underrepresented in the field of computing
- Link to your professional website (optional, but strongly recommended)
- Approximate cost of tuition and fees for one academic year
- Where and when you held an internship (if applicable)
- Three conferences you are most likely to attend
- Contact information for three references who are established researchers familiar with your research (at least one of which must be from your primary academic advisor/supervisor and only one letter can be from a current Microsoft employee). We highly encourage you to reach out to your references long before your proposal deadline. Microsoft will automatically provide instructions and request a reference letter from each of your three reference contacts separately as you submit your proposal. The sooner you submit your proposal, the more notice they will receive to upload a letter before the deadline. Those auto-generated emails will be sent from Microsoft Research Fellowship Program (firstname.lastname@example.org), which may end up in their spam folder. References will be asked to upload a letter in our online form. Note that all three contacts must submit your reference letters by Monday, September 30, 2019 at 5:00 PM Pacific Time in order for your proposal to be considered. Due to the number of submissions, we will not respond to questions asking if your references were submitted in time. You will receive an auto-generated confirmation email each time one of your references submits a letter. You may also log into your proposal to see if a letter has been received or not. It is your responsibility to follow up with your references.
- Proposals will be accepted via the online form in any of the following formats: Word document, text-only file, or PDF. Email or hard-copy submissions will not be considered.
- Proposals submitted to Microsoft will not be returned. Microsoft cannot assume responsibility for the confidentiality of information submitted in the proposal. Therefore, proposals should not contain information that is confidential, restricted, or sensitive. Microsoft reserves the right to make public the information on those proposals that receive awards, except those portions containing budgetary or personally identifiable information.
- Incomplete proposals will not be considered.
- Due to the volume of submissions, Microsoft Research cannot provide individual feedback on proposals that do not receive fellowship awards.
Below are the answers to frequently asked questions about the 2020 Microsoft Research PhD Fellowship.
Are international students eligible?
Yes, if you are a full-time international student attending a North American school.
What if I’m a student attending a university outside North America?
The Microsoft Research PhD Fellowship includes only North American schools. If you are a student attending a school outside North America, you are not eligible for this fellowship. Please check out our PhD Scholarship Programme in EMEA and our Fellowships at Microsoft Research Asia.
What if I am not starting my third year in academic year 2019–2020?
Students must be in their third year in a PhD program in the fall semester or quarter of 2019 to submit a proposal for this program. If you are starting your second year, you may be eligible for the Microsoft Research Ada Lovelace Fellowship. If you are in your fourth year or beyond of a PhD program, then you may be eligible for the Microsoft Research Dissertation Grant.
Do I have to be nominated by my university or can I nominate myself?
To be considered for the program, you must be nominated by your department within your university. If you are nominated, you will be contacted to submit a proposal.
Who is meant to submit the nomination? Does it have to be the Chair of the department or can it be my advisor?
The Department Chair’s office can designate any staff other than the student to submit the nomination. The nomination simply needs to be a coordinated effort with the Department Chair’s office. This is to ensure there are no more than three submissions from each department per fellowship and the Department Chair’s office is aware of who is submitting the three nominations.
Does the Department Chair have to submit a letter through the online tool as well?
No, we do not require a letter from the Department Chair.
Does whether or not a student's research is already being funded impact their eligibility for nomination and/or winning the award?
Not from our perspective. However, should the student be chosen for both a Microsoft Research fellowship and another industry fellowship, they will be asked to choose.
You specify very broad focus areas of research. Are there any proposals or projects that you are more interested in than others or is it up to us to choose? Is it mostly software solutions or is there any hardware interest?
There are plenty of both hardware and software projects currently in Microsoft Research. The reason the areas of research are broad is that Microsoft Research is very broad, and there are a number of people reviewing the fellowship proposals across a wide range of areas. Look at the work people in Microsoft Research are doing which will give you some idea of the focus areas within the broad areas to guide your focus area choice. In the end, propose the work you are interested in doing.
How do areas of interest factor into fellowship proposal evaluations? Are there areas of interest that Microsoft Research is more focused on this year?
It depends on the individuals involved in reviewing the proposal, and it is hard to say what is going to be of more interest. The trends of the industry are probably going to be reflected in what is interesting in general. Guiding question: Imagine you succeed. Tell us how someone’s life changes as a result.
How do I determine my primary and secondary research areas? Where is the appropriate place to describe how they relate to my work (whether it's methodologically or theoretically)?
- Your choices of primary and secondary areas help us choose who reviews your proposal.
- Pick areas that align with conferences/journals where you would publish.
- One-pager is the appropriate place to describe how research areas relate to research.
How do I choose which area to pick for my research area if my research is very interdisciplinary?
Microsoft Research is interdisciplinary, so it is something we understand. What you choose as a research area is a “soft” preference and will simply help us better route your proposal. Utilize the primary and secondary research area option to help capture and communicate your research area the best you can.
Here are some suggestions and guiding questions to help you choose a research area:
- Imagine you succeed. Tell us how someone’s life changes as a result.
- Do you have a home conference? Are there one or two conferences you go to in a more specific area?
- Is there a set of faculty/professors you know in a specific area?
- Who do you want to be reading your proposal?
- Who would you want to network with? What area of research are they in?
- Who would be most excited about my topic? What area of research are they in?
How related does my work need to be to Microsoft Research?
Your work should be of interest to researchers at Microsoft; however, it doesn’t need to directly line up with an existing project or topic. It is important for your work to be related enough that Microsoft researchers will be able to review it and have interest in supporting it. Microsoft Research is large, interdisciplinary, and covers a broad area — use the Research areas tab above as a guideline for the areas we cover. When in doubt, we suggest you browse the webpages of researchers who look like they may be related to your area and see if they have papers in the similar topics or publish in conferences you publish in and/or attend. If you find one or more such researchers that share these connections with you, then you can feel confident that your work is related enough to submit a proposal.
Thesis proposal or research statement
Are there specific pieces of information I am required to include in my research statement (i.e. research aims, timelines, deliverables) or are you looking for more of a narrative, descriptive format of a student's plan for their doctoral research?
Your research statement should be more of a narrative format. Timelines and deliverables are not necessary. We want to see what you are interested in, where your work is going, and how you would use this fellowship to further your research and contribute to the academic community.
What sort of balance is expected between what we have done, what we are doing, and what we are planning on doing?
When reviewing a proposal, we are looking for more of a future plan. Your research papers tell us what you have done, use the research statement to tell us where you are going.
If we have just submitted a paper that is going to guide a lot of direction going forward, would you recommend submitting that as preliminary data or attaching an unpublished paper to the proposal?
If it is relevant, and all co-authors approve of you submitting the unpublished work, we recommend including this in your submission. Again, all papers should be approved by all-co-authors, for both published and unpublished works.
Should the one-page research statement be a short synopsis of the five-page research statement, or should they contain distinctively different content?
The one-page research statement and the five-page research statement should not contain different content; one should be a shorter version of the other. The purpose of the shorter version is to help us triage where proposals go to get reviewed.
Should I be aware of formatting criteria while writing my research statement?
Your research statement should be no more than 5 pages including references with font no smaller than 10-point.
My undergrad was in a different field so the research was very different from what I am doing now. Who should write my reference letters?
Given you have three letters, it would be good to include a letter from one person who can speak about your current research and one person who has known you longer, even if it may not be in your current research area. The longer-term perspective is definitely important and valuable. The value of a letter is evaluating how you work, how you collaborate with people, and what your process is as a researcher. This transcends what your particular topic is. Keep in mind that one letter doesn’t have to address all things; across all three letters, we want to get a full picture of who you are over a longer term, but also insight into your recent work.
Are you more interested in learning about technical and research specific aspects of my work, or are other things, such as outreach/other university activities of interest as well?
The purpose of a letter of reference is to provide us with the bigger picture of what you are doing, how you work as a researcher, how you learn, how you approach projects, and how you collaborate with others. The letter will also provide us with insight from people who have been working with you and observing you for some amount of time.
It was suggested that recommendation letters come from established researchers. Is this limited to faculty members or would the inclusion of collaborators be acceptable as well?
At least one recommendation needs to come from an advisor, but letters of reference from collaborators are allowed. We are looking for people who can speak to you, your work as a researcher, and your character.
For the recommendation letters, is it a system where you list the people and your system will ask those people? Or do they have to send their recommendation letters to whomever is filling out the proposal?
Once you submit your proposal, those you provided as references will be sent an auto-generated email with instructions to upload their recommendation letters. The sooner you submit your proposal, the more notice they will receive to upload a letter before the deadline. We highly encourage you to reach out to your references long before your proposal deadline.
Who will review the proposals?
Proposals will be reviewed by researchers from Microsoft Research whose expertise covers a wide range of disciplines. After the first review, a selection of students will be invited for in-person interviews. Award recipients are chosen from the finalists.
When will I know the outcome of the review process?
Finalists will be contacted in November to book their travel for the interview. Due to the volume of submissions, Microsoft Research cannot provide individual feedback on proposals that do not receive fellowship awards.
How many proposals were there last year?
There were nearly 300 proposals submitted last year.
If selected, when will my fellowship begin?
Persons awarded a fellowship in January will receive their financial awards by September of that year. Microsoft sends payment directly to the university, who will disperse funds according to their guidelines.
Are there any tax implications for me if I receive this fellowship?
The tax implications for your tuition and fees and stipend are based on the policy at your university.
Will intellectual property be an issue if I am awarded a fellowship?
The Microsoft Research PhD Fellowship is not subject to any intellectual property (IP) restrictions unless and until the fellowship recipient also accepts an internship. If you accept an internship, you will be subject to the same restrictions as any other Microsoft intern.
Can I simultaneously receive fellowships from other companies?
If you accept a Microsoft Research PhD Fellowship, you may not receive another fellowship from another company or institution during the same academic period. Fellows accepting multiple fellowships will become ineligible to receive continued funding from Microsoft. Microsoft will at its sole discretion consider a joint fellowship with a government or non-profit organization. Please contact email@example.com for consideration.
Is childcare an approved use of my stipend?
Absolutely! There is no limit to the amount of your stipend that can be used for childcare.
Microsoft Research PhD Fellows
2019 PhD Fellows
Constantin investigates color centers in diamond and silicon carbide to utilize them as quantum bits and single photon sources for quantum information processing, which will enable exponential speed-up in a wide range of computing applications. To integrate color centers on a chip, he’s breaking new grounds in fabrication and design techniques to develop efficiently integrated photonics. Artificial intelligence and machine learning-based algorithms design the circuits of the future by engaging the full parameter space. To realize these designs using diamond and silicon carbide (materials that are new to photonics), Constantin developed fabrication methods in the Stanford Nanofabrication Facilities. He characterizes these devices at cryogenic temperatures in quantum optics labs and utilize them as spin-photon interfaces, or to generate nonclassical light. His current efforts are on developing large-scale quantum optical experiments involving several color centers entangled in a quantum circuit. Ultimately, he hopes to make progress toward the applications of color centers in universal quantum computers, quantum repeaters, and quantum transducers.
Rochester Institute of Technology
Danielle Gonzalez is currently a PhD student in the Center for Cybersecurity and Software Engineering Department at Rochester Institute of Technology (RIT). She also received a B.S. in Software Engineering from RIT in 2016. The goal of her research is to make security testing during development easier via unit testing. She uses static code analysis, large-scale data mining, natural language processing, and machine learning techniques to learn which critical paths and conditions in security tactic implementations can and should be unit-tested and automatically generate artifacts such as test case plans and recommendations that will make the testing process easier for developers while preserving their control over what is tested.
Carnegie Mellon University
Daehyeok Kim is a third year PhD student in the Computer Science Department at Carnegie Mellon University, where he is advised by Professor Srinivasan Seshan and Professor Vyas Sekar. His research interests lie in the intersection of systems and networking with a current focus on making data centers faster and more efficient by designing novel network primitives with advanced networking hardware such as programmable switches and RDMA NICs.
The University of Texas at Austin
Jayashree’s research aims at building the next generation of file systems: crash-consistent, energy-efficient, and IO-efficient. Crash consistency is the ability of a file system to ensure data and metadata consistency when a system crash occurs. It is important to build crash-consistent file systems, as applications rely on the guarantees provided by the underlying file system. In addition to crash-consistency, it is important for file systems to be energy-efficient because energy consumption is a key issue in both large-scale data centers as well as small hand-held devices. Furthermore, with the advent of storage technologies with limited write cycles, file systems have to be mindful of their IO footprint. Her work identifies the significance of each of these aspects, building tools and infrastructure to evaluate them, and understanding their impact on file system performance as a whole.
University of North Carolina at Chapel Hill
Ramakanth Pasunuru is a PhD student in Computer Science at University of North Carolina Chapel Hill, advised by Professor Mohit Bansal in the UNC-NLP group. His research focuses on developing knowledgeable language generation models that incorporate generalizable semantic skills via dynamic multi-task learning and multi-reward reinforcement learning methods, as well as multimodal conversational models that condition on video-based context and continually adapt via feedback and interaction. He is the recipient of an ACL 2017 Outstanding Paper Award and a COLING 2018 Area Chair Favorites Paper Award.
University of Wisconsin-Madison
As robot platforms become increasingly common in people’s homes and workplaces over the coming years, a central challenge will be ensuring people have effective ways to specify to a robot what they want it to do, and endowing robots with effective ways to communicate back to people such that the users can confidently interpret the robot’s intent and understanding given the command at hand. Daniel’s research has centered around this robot “specification-interpretability” problem, such as presenting remote telemanipulation paradigms that affords even novice users the ability to intuitively control a robot with minimal training, creating an automatic dynamic camera method that continuously optimizes a viewpoint for a remote user, and formulating a robot bimanual shared-control method inspired by how people naturally perform bimanual manipulations. Daniel is a PhD student at the University of Wisconsin-Madison, advised by Dr. Michael Gleicher and Dr. Bilge Mutlu.
Raghuvansh R. Saxena
Raghuvansh R. Saxena is a graduate student with the Department of Computer Science at Princeton University. His research involves developing and studying models of communication that capture the intricacies of modern communication systems, such as wireless and radio networks. Raghuvansh is also interested in other areas of theoretical computer science, such as algorithmic game theory and complexity theory. He received his Bachelor of Technology degree in Computer Science and Engineering from IIT Delhi in 2016, with Institute Rank 1.
Joana M. F. da Trindade
Massachusetts Institute of Technology
The goal of Joana’s research is to develop systems that enable data scientists and practitioners to analyze their data more efficiently and with less programming and/or cognitive effort. Despite the growing popularity and availability of graph data, existing graph analytics tools are either too slow, or too cumbersome to use; this forces upon biologists, chemists, and otherwise network scientists the choice between high-performance or usability. Existing graph databases provide accessible declarative language interfaces, yet their performance is nowhere near that of state-of-the-art graph processing libraries. Similarly, parallel graph processing libraries offer outstanding performance, but require expertise in parallel programming. So far in her PhD, Joana has identified several key challenges in this space, as well as contributed query optimization techniques that yield significant performance speedups for graph databases. Ultimately, she envisions a world where network data scientists can conduct their research without getting bogged down by cumbersome tools.
University of Maryland
Zuxuan is a third-year PhD student in the Department of Computer Science at the University of Maryland, advised by Professor Larry S. Davis. His research interests are computer vision and machine learning. In particular, his research focuses on developing efficient frameworks through conditional computation for automated visual understanding and learning robust feature representations with limited supervision. He is also interested in large-scale video recognition, and he has developed a few systems that achieved top-notch performance in several international benchmark competitions such as the YouTube-8M Video Classification Challenge.
Carnegie Mellon University
People have always used tools to extend the reach of their bodies and minds. Just as the invention of the lever enabled people to build monuments, the invention of pen and paper—and then the text editor—enabled people to synthesize ideas at a massively new scale. Today, computing power is cheap, networking ubiquitous, and data abundant. Yet today’s personal tools remain much the same as those of a few decades ago. Katherine’s research reimagines personal tools as modern mediums that natively incorporate computational intelligence—that is, techniques for modeling, search, and synthesis—to augment the human ability to think and create. Currently, Katherine leads the team that builds Penrose, a platform that enables people to create beautiful diagrams just by typing mathematical notation in plain text.
2018 PhD Fellows
As robots become more prevalent in our world, we’re challenged with programming them to learn faster and be more adaptable to different tasks and scenarios. Humans already have a high-level mastery of most of the tasks we’d like robots to learn. So Iretiayo is researching a way to map the signals from the human brain to useful reward signals through human-in-the-loop robot learning. This involves the development of techniques that use new modalities such as Brain-Computer Interfaces to allow humans to control and teach robots different skills, thus accelerating robot learning of complex tasks. To accomplish this, he is developing interactive robot learning systems that can leverage explicit human guidance through augmented reality platforms or implicit guidance signals through naturally evoked brain signals of a human to train the robot to perform different tasks.
Carnegie Mellon University
Virtually all modern machines, from our laptops to our phones, have multiple cores. It is thus of the utmost importance to be able to design correct and efficient concurrent programs. While there is ample work defining and proving the correctness of concurrent algorithms, surprisingly little is known about their efficiency. Many factors, like the number of processes available, the memory access patterns of the algorithm, and the cache coherency protocols of the machine, can greatly affect the performance of a given algorithm on a given machine. Naama’s research aims to formally and rigorously characterize the costs associated with concurrency and to provide a framework with which researchers can analyze the efficiency of the algorithms they develop.
Vincent Josua Hellendoorn
University of California, Davis
The world’s software is increasingly developed in globally distributed and highly decentralized settings, such as the open source domain. As a result, the main vessel for communication between developers is more often the code itself: its naming, style, and organization. Understanding this form of communication is necessary for tools that aim to intelligently support developers in creating software. Vincent’s research applies models that were originally designed for natural languages to source code to better understand the characteristics of real-world software. His work improves these (typically deep learning) models by integrating insights from software engineering research and programming language design. The resulting models are used to build tools that can assist developers in common, but taxing software engineering tasks like fault detection and API discovery.
University of Washington
Insect-scale aerial robots – the size of a penny – are tiny, flying bio-inspired platforms with sensing and computation abilities uniquely suited to a host of applications benefiting from their small scale and maneuverability including disaster relief, agriculture, search and rescue, and even dynamic 3D displays. Vikram’s research focuses on addressing the fundamental wireless networking and hardware challenges for these devices including power delivery, communication, and localization.
University of Washington
Natalie is a graduate student at the University of Washington studying physics with the Institute of Nuclear Theory under adviser Dr. Martin Savage. She has been inspired by the simulation of fundamental physics and has been intrigued by a new perspective using quantum-based computation—simulating many-body quantum systems with quantum systems. Excited by recent experimental digital quantum simulations of the vacuum properties of electrodynamics as well as the great success of non-perturbative techniques developed by the nuclear and particle physics communities over recent decades for studying the interactions of quarks and gluons, Natalie plans to spend her PhD research in the pursuit of simulating field theories on quantum computers. With this research, in concert with that of Microsoft Research, she hopes to unravel an answer to the basic question: is a quantum computer sufficient to efficiently simulate our natural world?
Machine learning algorithms are increasingly being used in socially consequential contexts, and numerous concerns have been raised with regard to the potential adverse impacts on disadvantaged populations. Manish’s research seeks to understand and develop techniques to mitigate such effects. This involves analyzing mechanisms that address particular societal biases as well as identifying fundamental trade-offs that exist between desirable properties of algorithmic systems. Ultimately, the aim of his research is to expose the normative choices necessary in algorithmic decision-making processes and provide insights as to the impacts these choices have.
Julia (Romanski) Gaudio
Massachusetts Institute of Technology
Julia’s research is in applied probability and algorithms. Currently, she is analyzing a model of interacting random walks. Julia initially proposed this model in order to study resource flow in an economic network, but there may be more applications of the model, for example in social science (e.g. modeling group interactions), or even in physics and chemistry. Her goal is to analyze the theoretical aspects of the model so that it can be used in applied settings. There are many theoretical questions to consider when studying the underlying Markov chain, such as phase transition, mixing time, estimation questions, and formulating an appropriate continuum version of the model.
University of Toronto
In the age of digital computing, we strive to communicate with computers in order to leverage their power to augment our human intelligence and creativity. We explore creative ideas by sketching out designs, manipulating models, and conversing with others. The computing power usually comes last, where the rich and dynamic ideas are downsampled to certain supported representations, by clicking commands from the menus or filling in numbers for the properties. Haijun’s research seeks to empower us to think with computers. He invents new representations of the content, coupled with intuitive interaction techniques, to enable flexible exploration and direct expression of ideas via graphical, gestural, and verbal communication that we are naturally capable of.
University of Utah
With the increasing presence of sensors, smartphones, and IoT projects, there has been an explosion in the amount of spatiotemporal data. As the data size grows beyond the storage and processing capacity of a single machine, we’re faced with the fundamental challenges of storing, processing, and analyzing large amounts of spatiotemporal data efficiently. Motivated by such observations, Dong’s research has led to building a distributed spatiotemporal data analytics engine called Simba, which features a simple yet expressive programming interface and high-performance query processing. While developing Simba as a full-fledged system, Dong explores both theory and system aspects by designing novel distributed algorithms for sophisticated spatiotemporal query and exploiting modern system techniques like full-stage code generation.
Justine Zhang is a PhD student in Information Science at Cornell University, advised by Professor Cristian Danescu-Niculescu-Mizil. Her research aims to develop models of conversations that explicitly encode the complex interactional dynamics they entail and are applicable to the diversity and scale of large online social systems. She received her Bachelor of Science in Computer Science from Stanford University.
2017 PhD Fellows
Michael B. Cohen
Massachusetts Institute of Technology
Michael’s research is focused on obtaining faster algorithms for fundamental linear algebra and optimization problems, generally with provable guarantees. Work includes dimensionality reduction—that is, transforming a high-dimensional problem into one of much lower dimension for faster processing. In addition, Michael is working on applying graph Laplacian solvers as a faster way to obtain max flow and minimum-cost flow outputs in graphic applications.
The Microsoft Research family is shocked and deeply saddened by the tragic loss of our colleague and friend Michael Cohen. Michael was a brilliant mathematician and a rising star in his field. He spent this past summer with us as a Microsoft Research Fellow, doing what he loved most. Over the summer, he made sweeping progress in online learning and online algorithms, two fields he had just recently become acquainted with. In addition to solving five open problems in these areas, he continued his substantial progress on the k-server problem, one of the most celebrated and notoriously difficult challenges in the space of adaptive algorithms.
Michael was a truly exceptional individual. We will remember Michael for his infectious smile and his larger-than-life personality. We will never forget his unrelenting curiosity, his thirst for knowledge, and his deep love for mathematics and theoretical foundations of computing. We are stunned by his loss. We will hold onto our memories of Michael, and know that his ideas and scientific accomplishments will continue on as important advances.
We extend our most sincere condolences to Michael’s family and friends.
Bita Darvish Rouhani
University of California, San Diego
To analyze massive and densely correlated data requires solving for two main challenges: resource constraints and complexity. One set of hurdles deals with the resource and/or application constraints such as real-time requirement, available energy, or memory. The other set of challenges arises due to the complexity of the underlying learning task, which requires more than traditional linear or polynomial analytical approaches to be sufficiently accurate. Bita’s research addresses the challenges associated with big data analytics using a combination of hardware-based and software-based approaches. Her work addresses these two critical aspects of big data scenarios by designing and building solutions and tools that are both data-aware and platform-aware.
Georgia Institute of Technology
Despite growing Internet access initiatives, political, economic, and social barriers continue to limit meaningful Internet engagement for individuals in resource-constrained communities. Michaelanne’s research explores how increasing access to the Internet influences lives and how one might use preexisting, local information infrastructures to design more effective services in emerging markets. Her current research is in Cuba, where, up until recently, Internet access was limited to 5 percent of the population. Through fieldwork, observation, and interviews, Michaelanne is developing a holistic understanding of how new Internet infrastructures interact with cultural values and local constraints. Her work explores how future Internet access initiatives in resource-constrained parts of the world might successfully map onto local information infrastructures.
University of Washington
Developing revenue maximization models that address uncertainty and specific real-world constraints requires new approaches and algorithms. Kira’s research looks to stretch revenue maximization models beyond restricted or known distributions, complex pricing mechanisms, and risk-neutrality, as well as apply them to nonrevenue (social good) outcomes. This research has applications ranging from blockchain-based systems, to kidney exchanges, routing traffic in a network, pricing package deliveries, selling electronics or any commerce application where auctions and pricing uncertainty exist.
The major successes in machine learning during the past few decades have been restricted to supervised settings where there is access to vast quantities of labeled data. In unlabeled, unsupervised contexts with significant uncertainty, machine learning has a long way to go. Aditya’s research focuses on designing and analyzing principled, efficient algorithms for probabilistic reasoning in unsupervised settings with unlabeled data, with a special emphasis on deep generative models that can scale to large and high-dimensional datasets, such as those found in energy conservation and weather forecasting applications.
University of Illinois at Urbana-Champaign
Silu’s research is around developing a collaborative data analytics program called DataHub. In essence, she is developing a multiuser system like GitHub, but for structured data. It features compact storage to handle large datasets as well as a rich query language for manipulations on data and versions. Early development goals include designing a compact storage engine with fast retrieval of versions, creating a unified query language for data provenance and versioning, and adapting relational databases to support versioning. This system is being built on top of PostgreSQL with a carefully designed data model and novel partitioning algorithms to efficiently support collaborative data management.
University of California, Los Angeles
The central motivation for Saswat’s research is to help programmers build reliable software with verified guarantees. Although several formal techniques have been developed for program verification, adoption has been low, primarily because two key requirements for these techniques—a formal specification and inductive invariants—still need to be provided manually, which is a cumbersome and error-prone process. Saswat’s research focuses on developing solutions that allow users to interactively generate and refine these requirements and thus help them formally verify their programs. His approach combines 1) data-driven insights from machine learning (on data collected from programs), 2) logical insights from formal reasoning (on program structure), and 3) effective user interaction models.
University of Michigan
The development community has designed many powerful, dynamic analyses that help programmers understand software, such as backward slicing, dynamic information flow tracking, memory checkers, invariant checkers, and data-race detectors. While these analyses are useful for debugging, developers use them only on extremely challenging bugs due to their high overhead. For example, it can take hours to calculate the backward slice of a complex program. Andrew’s research creates cluster-scale systems that allow developers to quickly understand and debug programs. The work parallelizes dynamic analyses across thousands of cores in a compute cluster, reducing analysis time from hours to seconds.
University of California, San Diego
Mengting focuses on building scalable machine learning algorithms to process massive and heterogeneous real-world human activity datasets. In particular, her research focuses on modeling human opinions and behavior to understand the dynamics of human activities and interactions. Specifically, she is working on two problems: the relationship between macro-knowledge and micro-opinions. Regarding the relationship between what someone knows in the general to how that informs their opinions or behaviors, she aims to effectively organize and summarize both factual knowledge and subjective opinions. Regarding the inverse, how micro-opinions relate to macro-knowledge, her work aims to interpret and predict behavior, studying the connection between efficient recommendation engines and established behavioral theories in social sciences. Mengting’s work has application in opinion-oriented question answering, recommendation systems, and e-commerce.
2016 PhD Fellows
|Salma Hosni Elmalaki
University of California, Los Angeles
North Carolina State University
Carnegie Mellon University
University of British Columbia, Vancouver
Georgia Institute of Technology
Massachusetts Institute of Technology
Georgia Institute of Technology
|Venkata Ravitheja Yelleswarapu
University of Pennsylvania
2015 PhD Fellows
|Lilian de Greef
University of Washington, Seattle
University of Massachusetts, Amherst
University of Waterloo
Massachusetts Institute of Technology
|Phitchaya Mangpo Phothilimthana
University of California, Berkeley
University of California, Berkeley
University of Waterloo
University of Washington, Seattle
University of California, Santa Barbara
2014 PhD Fellows
- Fadel Adib – Massachusetts Institute of Technology
- Yoav Artzi – University of Washington, Seattle
- Vijay Chidambaran – University of Wisconsin, Madison
- Richard Eisenberg – University of Pennsylvania
- Mayank Goel – University of Washington, Seattle
- Sergey Gorbunov – Massachusetts Institute of Technology
- Rishabh Iyer – University of Washington, Seattle
- Albert Ng – Stanford University
- Nihar Shah – University of California, Berkeley
- Abhinav Shrivastava – Carnegie Mellon University
- Niki Vazou – University of California, San Diego
- Yanqi Zhou – Princeton University
2013 PhD Fellows
- Bharath Hariharan – University of California, Berkeley
- Ahmed Kirmani – Massachusetts Institute of Technology
- Walter Lasecki – University of Rochester
- Michael Paul – Johns Hopkins University
- Gennady Pekhimenko – Carnegie Mellon University
- Mert Pilanci – University of California, Berkeley
- Jeffrey Rzeszotarski – Carnegie Mellon University
- Rahul Sharma – Stanford University
- Rashmi Vinayak – University of California, Berkeley
- Mathew Weinberg – Massachusetts Institute of Technology
- Mingchen Zhao – University of Pennsylvania
- Yufei Zhao – Massachusetts Institute of Technology
2012 PhD Fellows
- Utku Ozan Candogan – Massachusetts Institute of Technology
- Gabriel Cohn – University of Washington, Seattle
- George Dahl – University of Toronto
- Sigal Oren – Cornell University
- Ashish Patro – University of Wisconsin, Madison
- Russell Power – New York University
- Franziska Roesner – University of Washington, Seattle
- Michael Rubinstein – Massachusetts Institute of Technology
- Julia Schwarz – Carnegie Mellon University
- Rishabh Singh – Massachusetts Institute of Technology
- Richard Socher – Stanford University
- Bangpeng Yao – Stanford University
2011 PhD Fellows
- Michael Bernstein – Massachusetts Institute of Technology
- Morgan Dixon – University of Washington, Seattle
- Matthew Fredrikson – University of Wisconsin, Madison
- Allison Lewko – University of Texas, Austin
- Renato Paes Leme – Cornell University
- Qiang Liu – University of California, Irvine
- Richard Peng – Carnegie Mellon University
- Lenin Ravindranath Sivalingam – Massachusetts Institute of Technology
- Trang Thai – Georgia Institute of Technology
- Yuandong Tian – Carnegie Mellon University
- Oriol Vinyals – University of California, Berkeley
- Chi Wang – University of Illinois, Urbana-Champaign
2010 PhD Fellows
- Silas Boyd-Wickizer – Massachusetts Institute of Technology
- David Erickson – Stanford University
- Charles Han – Columbia University
- William Harris – University of Wisconsin, Madison
- Chris Harrison – Carnegie Mellon University
- Dilip Krishnan – New York University
- Jian Peng – Toyota Technological Institute at Chicago
- Shubhangi Saraf – Massachusetts Institute of Technology
- Dafna Shahaf – Carnegie Mellon University
- Jonah Sherman – University of California, Berkeley
2009 PhD Fellows
- Alekh Agarwal – University of California, Berkeley
- Timothy Austin – University of California, Los Angeles
- Katherine Coons – University of Texas, Austin
- Lakshmi Ganesh – Cornell University
- Jean-Francois Lalonde – Carnegie Mellon University
- Edith Law – Carnegie Mellon University
- Christopher Le Dantec – Georgia Institute of Technology
- Huijia Lin – Cornell University
- Kayur Patel – University of Washington
- Snehit Prabhu – Columbia University
- Shravan Rayanchu – University of Wisconsin
- Sven Seuken – Harvard University
- Yaron Singer – University of California, Berkeley
- Ross Tate – University of California, San Diego
- Xiang Zhang – University of North Carolina, Chapel Hill
2008 PhD Fellows
- Aruna Balasubramanian – University of Massachusetts
- Pravin Bhat – University of Washington
- Michael Carbin – Massachusetts Institute of Technology
- Eric S. Chung – Carnegie Mellon University
- Prabal Dutta – University of California, Berkeley
- Jon Froehlich – University of Washington
- Vipul Goyal – University of California, Los Angeles
- Nitin Gupta – Cornell University
- Sasa Junuzovic – University of North Carolina at Chapel Hill
- Ece Semiha Kamar – Harvard University
- Dave Levin – University of Maryland
- Rohan Narayana Murty – Harvard University
- Zvonimir Rakamaric – University of British Columbia
- T. Scott Saponas – University of Washington
- John Wright – University of Illinois
- Yisong Yue – Cornell University
2008 Social Impact Award
- Andrea Grimes – Georgia Institute of Technology