Jon McAuliffe
Associate Professor (Adjunct)
Statistics Department
University of California, Berkeley
449 Evans Hall
Berkeley, CA 94720



I received a Ph.D. in statistics here at Berkeley. My thesis advisor was Professor Michael Jordan.

Academic CV: pdf

My areas of interest include

  • statistical prediction and machine learning
  • large-scale statistical inference
  • kernel methods
  • nonparametric and semiparametric estimation
  • bioinformatics and computational biology
  • prediction and optimization problems in finance

Professional service

  • Area chair, Conference on Neural Information Processing Systems (NIPS), 2009.
  • Area chair, International Conference on Artificial Intelligence and Statistics (AISTATS), 2009.
  • IMS invited session organizer, Joint Statistical Meetings (JSM), 2007. Session title: Machine learning and optimization.

Publications and slides

Please email me if you want a pdf of my article but do not have electronic access to the journal. Where permitted, I will supply you with a postprint.


Michael Braun and Jon D. McAuliffe.
“Variational inference for large-scale models of discrete choice.”
Journal of the American Statistical Association 105(489), March 2010: 324-335.

Sanne Nygaard, Alexander Braunstein, Gareth Malsen, Stijn van Dongen, Paul P. Gardner, Anders Krogh, Thomas D. Otto, Arnab Pain, Matthew Berriman, Jon McAuliffe, Emmanouil T. Dermitzakis, and Daniel C. Jeffares.
“Long- and short-term selective foreces on malaria parasite genomes.”
PLoS Genetics 6(9), September 2010: e1001099.


Alexander Braunstein, Zhi Wei, Shane Jensen, and Jon McAuliffe.
“A spatially varying two-sample recombinant coalescent, with applications to HIV escape response.”
Advances in Neural Information Processing Systems 21 (NIPS 2008).


S. M. Sweeney, J. P. Orgel, A. Fertala, J. D. McAuliffe, K. R. Turner, G. A. Di Lullo, S. Chen, O. Antipova, S. Perumal, L. Ala-Kokko, A. Forlino, W. A. Cabral, A. M. Barnes, J. C. Marini, and J. D. San Antonio.
“Candidate cell and matrix interaction domains on the collagen fibril, the predominant protein of vertebrates.”
Journal of Biological Chemistry 283(30): 21187-21197.

David M. Blei and Jon D. McAuliffe.
“Supervised topic models.”
Advances in Neural Information Processing Systems 20 (NIPS 2007).


“Machine learning and modern biological data.”
Introductory Overview Lecture, Joint Statistical Meetings, Salt Lake City.
Slides: pdf


Peter L. Bartlett, Michael I. Jordan, and Jon D. McAuliffe.
“Discussion of ‘Support vector machines with applications’.”
Statistical Science 21(3), August 2006: 341-346.

Jon D. McAuliffe, David M. Blei, and Michael I. Jordan.
“Nonparametric empirical Bayes for the Dirichlet process mixture model.”
Statistics and Computing 16(1), March 2006: 5-14.

Peter L. Bartlett, Michael I. Jordan, and Jon D. McAuliffe.
“Convexity, classification, and risk bounds.”
Journal of the American Statistical Association 101(473), March 2006: 138-156.


Jon D. McAuliffe, Michael I. Jordan, and Lior Pachter. “Subtree power analysis and species selection for comparative genomics.”
Proc. of the National Academy of Sciences 102(22), 31 May 2005: 7900-5.

Prasad Gyaneshwar, Oleg Paliy, Jon McAuliffe, Adriane Jones, Michael I. Jordan, and Sydney Kustu. “Lessons from Escherichia coli genes similarly regulated in response to nitrogen and sulfur limitation.”
Proc. of the National Academy of Sciences 102(9), 1 Mar 2005: 3453-8.

Prasad Gyaneshwar, Oleg Paliy, Jon McAuliffe, David L. Popham, Michael I. Jordan, and Sydney Kustu. “Sulfur and nitrogen limitation in Escherichia coli K-12: specific homeostatic responses.”
Journal of Bacteriology 187(3), Feb 2005: 1074-1090.


Jon McAuliffe, Lior Pachter, and Michael I. Jordan. “Multiple-sequence functional annotation and the generalized hidden Markov phylogeny.”
Bioinformatics 20(12), 12 Aug 2004: 1850-60.

Peter L. Bartlett, Michael I. Jordan, and Jon D. McAuliffe. “Discussion of boosting papers.” Annals of Statistics 32(1), Feb 2004: 85-91.


Benjamin I. P. Rubinstein, Jon McAuliffe, Simon Cawley, Marimuthu Palaniswami, Kotagiri Ramamohanarao, and Terence P. Speed.
“Machine learning in low-level microarray analysis.”
SIGKDD Explorations 5(2), Dec 2003: 130-9.

Rebecca W. Corbin, Oleg Paliy, Feng Yang, Jeffrey Shabanowitz, Mark Platt, Charles E. Lyons, Jr., Karen Root, Jon McAuliffe, Michael I. Jordan, Sydney Kustu, Eric Soupene, and Donald F. Hunt. “Toward a protein profile of Escherichia coli: comparison to its transcription profile.”
Proc. of the National Academy of Sciences 100(16), 5 Aug 2003: 9232-7.

Peter L. Bartlett, Michael I. Jordan, and Jon D. McAuliffe.
“Large margin classifiers: convex loss, low noise, and convergence rates.”
Advances in Neural Information Processing Systems 16 (NIPS 2003).

Dario Boffelli, Jon McAuliffe, Dmitriy Ovcharenko, Keith D. Lewis, Ivan Ovcharenko, Lior Pachter, and Edward M. Rubin. “Phylogenetic shadowing of primate sequences to find functional regions of the human genome.”
Science 299(5611), 28 Feb 2003: 1391-4.
pdf | online supplement | Perspective


From 1995 to 1999 I worked at D. E. Shaw & Co., where among other things I developed statistical equity arbitrage trading systems.

For a year thereafter I researched statistical recommender systems at; the (now defunct) Purchase Circle “uniquely popular” community-based item rankings are an example.

I spent the summer before graduate school at the mobile-computing startup Vindigo, researching

  • MDL-based compression approaches for handheld software
  • graph algorithms for dynamically generated navigational directions.
Vindigo was an amazing mobile (Palm) application in 2000, and it had a colorful history thereafter — here is what finally happened.

During graduate school, I spent a summer at Affymetrix researching genotype-calling methods for the original SNP chip.

Also during graduate school, I worked with Efficient Frontier on click-through rate prediction for keyword (search-result) ads.


Savage’s approach to research, via Mosteller:

  1. As soon as a problem is stated, start right away to solve it. Use simple examples.
  2. Keep starting from first principles, explaining again and again what you are trying to do.
  3. Believe that this problem can be solved and that you will enjoy working it out.
  4. Don’t be hampered by the original problem statement. Try other problems in its neighborhood; maybe there’s a better problem than yours.
  5. Work an hour or so on it frequently.
  6. Talk about it; explain it to people.

Quotes worth quoting:

  • Good judgment comes from experience. Experience comes from bad judgment.
    —Jim Horning

  • Dealing with failure is easy: work hard to improve. Success is also easy to handle: you’ve solved the wrong problem. Work hard to improve.
    —Alan J. Perlis

  • However beautiful the strategy, you should occasionally look at the results.
    —Winston Churchill

  • I have yet to see any problem, however complicated, which, when looked at in the right way, did not become still more complicated.
    —Poul Anderson

  • The difference between theory and practice: in theory, there’s no difference between theory and practice; in practice, there is.
    —Jan L. A. van de Snepscheut

  • The most exciting phrase to hear in science is not “Eureka!” but “That’s funny...”
    —Isaac Asimov

  • Don’t worry about people stealing your ideas. If your ideas are any good, you’ll have to ram them down people’s throats.
    —Howard Aiken

  • The secret to creativity is knowing how to hide your sources.
    —Albert Einstein

  • Men never do evil so cheerfully and completely as when they do it from religious conviction.
    —Blaise Pascal

  • I was unable to find flaws in my ‘proof’ for quite a while, even though the error is very obvious. It was a psychological problem, a blindness, an excitement, an inhibition of reasoning by an underlying fear of being wrong. Techniques leading to the abandonment of such inhibitions should be cultivated by every honest mathematician.
    —John R. Stallings Jr. [on his false proof of Poincare’s conjecture]

  • For sheer brilliance I could divide all those whom I have taught into two groups: one contained a single outstanding boy, R. A. Fisher; the other all the rest.
    —Arthur Vassal, Fisher's biology teacher at Harrow

  • [Fisher] fitted the classical definition of a gentleman: he never insulted anyone unintentionally.
    —J.F. Crow

  • I occasionally meet geneticists who ask me whether it is true that the great geneticist R. A. Fisher was also an important statistician.
    —L. J. Savage

  • If the topic of regression comes up in a trial, the side that must explain regression to the jury will lose.
    —David A. Freedman

I don’t know Web design, but I know someone who knows it.

Last updated: 1 Feb 2013