Daniel Roy

news

JAN 2011
I am delighted to report that I have been elected as a Junior Research Fellow at Emmanuel College, University of Cambridge.

SEP 2010
I have accepted a Newton International Fellowship at the University of Cambridge. I will be joining Zoubin Ghahramani as a member of the Machine Learning Group.

research

My research interests lie at the intersection of computer science, statistics and probability theory; I study probabilistic programming languages to develop computational perspectives on fundamental ideas in probability theory and statistics. I am particularly interested in the use of recursion to define nonparametric distributions on data structures; representation theorems that connect computability and probabilistic structures; and the complexity of inference.

NIPS workshop

I co-organized a workshop on probabilistic programming for statistics and machine learning at NIPS*2008 (with Vikash Mansinghka, John Winn, David McAllester and Josh Tenenbaum).

teaching

6.437 Inference and Information
with Polina Golland and Greg Wornell (Spring 2008)

6.867 Machine Learning
with Tommi Jaakkola (Fall 2007)

6.035 Computer Language Engineering
with Martin Rinard and Saman Amarasinghe (Fall 2003)

academic

Hello, my name is Dan(iel) and I am a graduate student in the EECS PhD program in computer science at the Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory (CSAIL). I collaborate with my advisor Leslie Kaelbling as well as members of Josh Tenenbaum's Computational Cognitive Science group on work at the intersection of theoretical computer science, cognitive science and machine learning. My other interests include scalable AI, algorithmic complexity, logical representations, grounding, coding and information theory, automated software synthesis, and programming language theory.

personal

I enjoy many outdoor activities and sports such as skiing, hiking, barefoot running, cycling, volleyball, basketball and rowing. I also enjoy dancing salsa, studying cello and traveling. I used to compete in student film competitions and produce electronic music, though I've fallen out of practice. I am a resident tutor at Leverett House at Harvard University, where I advise undergraduates in Computer Science and related subjects. In addition to participating in the Senior Common Room, I row crew and play football, soccer and many other intramural sports for Leverett.

am i hot?

Am I Hot? In my freshman year at MIT, I created a website called AmIHot.com. After it was featured on Howard Stern's national radio show in 2000, its popularity sky-rocketed. I took Spring term of my sophomore year at MIT off to work on AmIHot.com full time and make it profitable. Before I sold AmIHot in 2004 to HotOrNot.com, our web servers were serving up millions of page views, handled by tens of thousands of lines of code running on multiple database and web servers.

volleyball

I played "opposite" on the MIT Men's Varsity Volleyball Team. I was captain of the Cambridge University Men's Volleyball team during my year abroad at Cambridge during the 2001-2002 season, the best season in its history. We won both the English Volleyball Association (EVA) championship and the British Universities Sports Association (BUSA) championionship, a feat not achieved by any British university in the previous decade. The BUSA win was Cambridge University's third ever (and first for volleyball). This earned us a spot at the European University Championships in Athens, Greece, where we came in 7th in Europe. I personally set the season record for most points in a season (kills, aces and blocks). In honor of our hard work, the entire starting team was awarded "Full Blues," a distinction reserved for Cambridge's top athletes. (This was a long time ago, but I'm still very proud of our achievements.)

Source: Jesus College Virtual Tour

study abroad

I spent my junior year abroad under the Cambridge-MIT Exchange program. CMI/CME is a great program and I highly recommend it. If you are interested in participating, I am more than willing to discuss my experience.

valid xhtml

Valid XHTML 1.1 re-validate
Valid CSS /styles/css1, /styles/css2

preprints (last updated: 05/10)

On the computability of conditional probability
(with Nate Ackerman and Cameron Freer)
[ arXiv:1005.3014 ]

Computable de Finetti measures
(with Cameron Freer)
[ arXiv:0912.1072, PDF ]

publications

Noncomputable conditional distributions
(with Nate Ackerman and Cameron Freer)
Proc. Logic in Computer Science (LICS), 2011.

Bayesian Policy Search with Policy Priors
David Wingate, Leslie P. Kaelbling, Daniel M. Roy, Noah D. Goodman, and Joshua B. Tenenbaum
Proc. Int. Joint Conf. on Artificial Intelligience (IJCAI), 2011.

When are probabilistic programs probably computationally tractable?
(with Cameron Freer and Vikash Mansinghka)
NIPS Workshop on Monte Carlo Methods for Modern Applications, 2010.
[ PDF bibtex ]

Posterior distributions are computable from predictive distributions
(with Cameron Freer)
Proc. Artificial Intelligence and Statistics (AISTATS), 2010.
[ PDF bibtex ]

Complexity of Inference in Topic Models
David Sontag and Daniel Roy
NIPS Workshop on Applications for Topic Models: Text and Beyond, 2009.
[ PDF bibtex ]

The Infinite Latent Events Model
David Wingate, Noah D. Goodman, Daniel M. Roy, and Joshua B. Tenenbaum
Proc. Uncertainty in Artificial Intelligence (UAI), 2009.
[ PDF bibtex ]

Computable exchangeable sequences have computable de Finetti measures
(with Cameron Freer)
Proc. Computability in Europe (CiE), 2009.
[ PDF bibtex ]

Exact and Approximate Sampling by Systematic Stochastic Search
Vikash Mansinghka, Daniel M. Roy, Eric Jonas, and Joshua Tenenbaum
Proc. Artificial Intelligence and Statistics (AISTATS), 2009.
[ PDF bibtex ]

The Mondrian Process
(with Yee Whye Teh)
Adv. Neural Information Processing Systems 21 (NIPS), 2009.
[ PDF bibtex ]

Video animation of the Mondrian process as one zooms into the origin (under a beta Levy rate measure at time t=1.0). See also the time evolution of a Mondrian process on the plane as we zoom in with rate proportional to time. In both cases, the colors are chosen at random from a palette. These animations were produced by Yee Whye in Matlab. For now, we reserve copyright, but please email me and we'll be more than likely happy to let you use them.

A stochastic programming perspective on nonparametric Bayes
Daniel M. Roy, Vikash Mansinghka, Noah Goodman, and Joshua Tenenbaum
ICML Workshop on Nonparametric Bayesian, 2008.

Church: a language for generative models
Noah Goodman, Vikash Mansinghka, Daniel M. Roy, Keith Bonawitz, and Joshua Tenenbaum
Proc. Uncertainty in Artificial Intelligence (UAI), 2008.
[ PDF bibtex ]

Bayesian Agglomerative Clustering with Coalescents
Yee Whye Teh, Hal Daumé III, and Daniel M. Roy
Adv. Neural Information Processing Systems 20 (NIPS), 2008.
[ arXiv:0907.0781, PDF bibtex ]

Discovering Syntactic Hierarchies
Virginia Savova, Daniel M. Roy, Lauren Schmidt, and Joshua B. Tenenbaum
Proc. Cognitive Science (COGSCI), 2007.
[ PDF bibtex ]

AClass: An online algorithm for generative classification
Vikash K. Mansinghka, Daniel M. Roy, Ryan Rifkin, and Joshua B. Tenenbaum
Proc. Artificial Intelligence and Statistics (AISTATS), 2007.
[ PDF bibtex ]

Efficient Bayesian Task-level Transfer Learning
Daniel M. Roy and Leslie P. Kaelbling
Proc. Int. Joint Conf. on Artificial Intelligience (IJCAI), 2007.
[ PDF bibtex ]

Learning Annotated Hierarchies from Relational Data
Daniel M. Roy, Charles Kemp, Vikash Mansinghka, and Joshua B. Tenenbaum
Adv. Neural Information Processing Systems 19 (NIPS), 2007.
[ PDF bibtex ]

Clustered Naive Bayes
MEng thesis, Massachusetts Institute of Technology, 2006.
[ PDF bibtex ]

Enhancing Server Availability and Security Through Failure-Oblivious Computing
Martin Rinard, Cristian Cadar, Daniel Dumitran, Daniel M. Roy, Tudor Leu, and William S. Beebee, Jr.
Proc. Operating Systems Design and Implementation (OSDI), 2004.
[ PDF bibtex ]

A Dynamic Technique for Eliminating Buffer Overflow Vulnerabilities (and Other Memory Errors)
Martin Rinard, Cristian Cadar, Daniel Dumitran, Daniel M. Roy, and Tudor Leu
Proc. Annual Computer Security Applications Conference (ACSAC), 2004.
[ PDF bibtex ]

Efficient Specification-Assisted Error Localization
Brian Demsky, Cristian Cadar, Daniel M. Roy, and Martin C. Rinard
Proc. Workshop on Dynamic Analysis (WODA), 2004.
[ PDF bibtex ]

Efficient Specification-Assisted Error Localization and Correction
Brian Demsky, Cristian Cadar, Daniel M. Roy, and Martin C. Rinard
MIT CSAIL Technical Report 927. November, 2003.
[ PDF bibtex ]