Who Am I?

I am currently an Assistant Professor in the Department of Electrical Engineering and Computer Science at the University of Michigan.
My research focus is Machine Learning, and I like discovering connections between Optimization, Statistics, and Economics. I finished my PhD at UC Berkeley with Peter Bartlett in 2011, and I was a Simons postdoctoral fellow with Michael Kearns for the following two years.

News and Links:

Recent Talks:

My Excellent Research Group at the Univ. of Michigan

Publications

Conference and Workshop Papers

  • Low-cost learning via active data procurement
    [PDF] -- J.A., Yiling Chen, Chien-Ju Ho, Bo Waggoner
    /2015/ -- Proceedings of the Sixteenth ACM Conference on Economics and Computation
  • Financialized methods for market-based multi-sensor fusion
    [PDF] -- J.A., M. Johnson-Roberson
    /2015/ -- IEEE/RSJ International Conference on Intelligent Robots and Systems
  • Price Discovery in Subgradient Combinatorial Auctions
    [PDF] -- J.A., S\'ebastien Lahaie, Matus Telgarsky
    /2015/ -- Auctions, Market Mechanisms, and Their Applications
  • A Market Framework for Eliciting Private Data
    [N/A] -- Bo Waggoner, Rafael Frongillo, J.A.
    /2015/ -- Advances in Neural Information Processing Systems 28
  • Fighting Bandits with a New Kind of Smoothness
    [N/A] -- J.A., Chansoo Lee, Ambuj Tewari
    /2015/ -- Advances in Neural Information Processing Systems 28
  • Jamming Defense Against a Resource-Replenishing Adversary in Multi-channel Wireless Systems
    [PDF] -- Qingsi Wang, Shang-Pin Sheng, J.A., Mingyan Liu
    /2014/ -- (WiOpt) International Symposium on Modeling and Optimization in Mobile, Ad Hoc and Wireless Networks
  • Information aggregation in exponential family markets
    [PDF] -- J.A., Sindhu Kutty, S{\'e}bastien Lahaie, Rahul Sami
    /2014/ -- (EC) Proceedings of the 15th ACM conference on Economics and computation
  • A general volume-parameterized market making framework
    [PDF] -- J.A., Rafael M Frongillo, Xiaolong Li, Jennifer Wortman Vaughan
    /2014/ -- (EC) Proceedings of the 15th ACM conference on Economics and computation
  • Online Linear Optimization via Smoothing
    [PDF] -- J.A., Chansoo Lee, Abhinav Sinha, Ambuj Tewari
    /2014/ -- (COLT) Conference on Learning Theory
  • How to Hedge an Option Against an Adversary: Black-Scholes Pricing is Minimax Optimal
    [PDF] -- J.A., Peter Bartlett, Rafael Frongillo, Andre Wibisono
    /2013/ -- (NIPS) Advances in Neural Information Processing Systems
  • Adaptive Market Making via Online Learning
    [PDF] -- J.A., Satyen Kale
    /2013/ -- (NIPS) Advances in Neural Information Processing Systems
  • Minimax optimal algorithms for unconstrained linear optimization
    [PDF] -- Brendan McMahan, J.A.
    /2013/ -- (NIPS) Advances in Neural Information Processing Systems
  • Large-Scale Bandit Problems and $\{$KWIK$\}$ Learning
    [PDF] -- J.A., Kareem Amin, Moez Draief, Michael Kearns
    /2013/ -- (ICML) Proceedings of the 30th International Conference on Machine Learning
  • Minimax option pricing meets black-scholes in the limit
    [PDF] -- J.A., Rafael M Frongillo, Andre Wibisono
    /2012/ -- (STOC) Proceedings of the 44th Symposium on Theory of Computing
  • A Characterization of Scoring Rules for Linear Properties.
    [PDF] -- J.A., Rafael M Frongillo
    /2012/ -- (COLT) Conference on Learning Theory
  • A Collaborative Mechanism for Crowdsourcing Prediction Problems.
    [PDF] -- J.A., Rafael M Frongillo
    /2011/ -- (NIPS) Advances in Neural Information Processing Systems
  • An optimization-based framework for automated market-making
    [PDF] -- J.A., Yiling Chen, Jennifer Wortman Vaughan
    /2011/ -- (EC) Proceedings of the 12th ACM conference on Electronic commerce
  • Blackwell approachability and low-regret learning are equivalent
    [PDF] -- J.A., Peter L Bartlett, Elad Hazan
    /2011/ -- (COLT) Conference on Learning Theory
  • Repeated Games against Budgeted Adversaries.
    [PDF] -- J.A., Manfred K Warmuth
    /2010/ -- (NIPS) Advances in Neural Information Processing Systems
  • A regularization approach to metrical task systems
    [PDF] -- J.A., Peter L Bartlett, Niv Buchbinder, Isabelle Stanton
    /2010/ -- (ALT) Algorithmic Learning Theory
  • A stochastic view of optimal regret through minimax duality
    [PDF] -- J.A., Alekh Agarwal, Peter L Bartlett, Alexander Rakhlin
    /2009/ -- (COLT) Conference on Learning Theory
  • Beating the adaptive bandit with high probability
    [PDF] -- J.A., Alexander Rakhlin
    /2009/ -- (COLT) Conference on Learning Theory
  • Optimal strategies from random walks
    [PDF] -- J.A., Manfred K Warmuth, Joel Yellin
    /2008/ -- (COLT) Conference on Learning Theory
  • Competing in the Dark: An Efficient Algorithm for Bandit Linear Optimization.
    [PDF] -- J.A., Elad Hazan, Alexander Rakhlin
    /2008/ -- (COLT) Conference on Learning Theory
    COLT 2008 Best Student Paper Award & 2008 Pat Goldberg Memorial Best Paper Award from IBM
  • Optimal strategies and minimax lower bounds for online convex games
    [PDF] -- J.A., Peter L Bartlett, Alexander Rakhlin, Ambuj Tewari
    /2008/ -- (COLT) Conference on Learning Theory
  • Web spam identification through content and hyperlinks
    [PDF] -- J.A., Olivier Chapelle, Carlos Castillo
    /2008/ -- (AirWeb) Proceedings of the 4th international workshop on Adversarial information retrieval on the web
  • Online discovery of similarity mappings
    [PDF] -- Alexander Rakhlin, J.A., Peter L Bartlett
    /2007/ -- (ICML) Proceedings of the 24th international conference on Machine learning
  • Multitask learning with expert advice
    [PDF] -- J.A., Peter Bartlett, Alexander Rakhlin
    /2007/ -- (COLT) Conference on Learning Theory
  • Continuous experts and the binning algorithm
    [PDF] -- J.A., John Langford, Manfred K Warmuth
    /2006/ -- (COLT) Conference on Learning Theory

Journal Papers

  • Efficient market making via convex optimization, and a connection to online learning
    [PDF] -- J.A., Yiling Chen, Jennifer Wortman Vaughan
    /2013/ -- ACM Transactions on Economics and Computation
  • Interior-point methods for full-information and bandit online learning
    [PDF] -- J.A., Elad Hazan, Alexander Rakhlin
    /2012/ -- IEEE Transactions on Information Theory
  • Graph regularization methods for web spam detection
    [PDF] -- J.A., Olivier Chapelle, Carlos Castillo
    /2010/ -- Machine Learning Journal
  • A New Approach to Collaborative Filtering: Operator Estimation with Spectral Regularization
    [PDF] -- J.A., Francis Bach, Theodoros Evgeniou, Jean-Philippe Vert
    /2009/ -- The Journal of Machine Learning Research
  • Eliciting consumer preferences using robust adaptive choice questionnaires
    [PDF] -- J.A., Theodoros Evgeniou, Olivier Toubia, Jean-Philippe Vert
    /2008/ -- IEEE Transactions on Knowledge and Data Engineering

Technical Reports on the ArXiv

  • Actively Purchasing Data for Learning
    [PDF] -- J.A., Yiling Chen, Chien{-}Ju Ho, Bo Waggoner
    /2015/ -- CoRR, ArXiv ID: abs/1502.05774
  • Spectral Smoothing via Random Matrix Perturbations
    [PDF] -- J.A., Chansoo Lee, Ambuj Tewari
    /2015/ -- CoRR, ArXiv ID: abs/1507.03032
  • Faster Convex Optimization: Simulated Annealing with an Efficient Universal Barrier
    [PDF] -- J.A., Elad Hazan
    /2015/ -- CoRR, ArXiv ID: abs/1507.02528
  • Online Linear Optimization via Smoothing
    [PDF] -- J.A., Chansoo Lee, Abhinav Sinha, Ambuj Tewari
    /2014/ -- CoRR, ArXiv ID: abs/1405.6076
  • Information Aggregation in Exponential Family Markets
    [PDF] -- J.A., Sindhu Kutty, S{\'e}bastien Lahaie, Rahul Sami
    /2014/ -- CoRR, ArXiv ID: abs/1402.5458
  • Minimax Option Pricing Meets Black-Scholes in the Limit
    [PDF] -- J.A., Rafael M. Frongillo, Andre Wibisono
    /2012/ -- CoRR, ArXiv ID: abs/1202.2585
  • A Collaborative Mechanism for Crowdsourcing Prediction Problems
    [PDF] -- J.A., Rafael M. Frongillo
    /2011/ -- CoRR, ArXiv ID: abs/1111.2664
  • An Optimization-Based Framework for Automated Market-Making
    [PDF] -- J.A., Yiling Chen, Jennifer Wortman Vaughan
    /2010/ -- CoRR, ArXiv ID: abs/1011.1941
  • Blackwell Approachability and Low-Regret Learning are Equivalent
    [PDF] -- J.A., Peter L. Bartlett, Elad Hazan
    /2010/ -- CoRR, ArXiv ID: abs/1011.1936
  • A Stochastic View of Optimal Regret through Minimax Duality
    [PDF] -- J.A., Alekh Agarwal, Peter L. Bartlett, Alexander Rakhlin
    /2009/ -- CoRR, ArXiv ID: abs/0903.5328
  • A New Approach to Collaborative Filtering: Operator Estimation with Spectral Regularization
    [PDF] -- Francis Bach, J.A., Jean-Philippe Vert, Theodoros Evgeniou
    /2008/ -- CoRR, ArXiv ID: abs/0802.1430
  • Low-rank matrix factorization with attributes
    [PDF] -- J.A., Francis Bach, Theodoros Evgeniou, Jean-Philippe Vert
    /2006/ -- CoRR, ArXiv ID: abs/cs/0611124

Open Problems

  • Does an Efficient Calibrated Forecasting Strategy Exist?
    [PDF] -- J.A., Shie Mannor
    /2011/ -- (COLT) Conference on Learning Theory
  • Can We Learn to Gamble Efficiently?
    [PDF] -- J.A.
    /2010/ -- (COLT) Conference on Learning Theory
  • Minimax Games with Bandits.
    [PDF] -- J.A., Manfred K Warmuth
    /2009/ -- (COLT) Conference on Learning Theory
  • An Efficient Bandit Algorithm for root-T-Regret in Online Multiclass Prediction?
    [PDF] -- J.A., Alexander Rakhlin
    /2009/ -- (COLT) Conference on Learning Theory

Talks and Lectures on Video

  • Princeton Theory Lunch, Oct. 26, 2012: Minimax Option Pricing Meets Black-Scholes in the Limit
    Video at The Center for Computational Intractability
  • NIPS 2011: A Collaborative Mechanism for Crowdsourcing Prediction Problems
  • COLT 2011: Blackwell Approachability and No-Regret Learning are Equivalent
  • ICML 2012 Tutorial: Prediction, Belief, and Markets
    Video on TechTalks.Tv (poor sound quality)
  • Google TechTalk 2007: Prediction, Belief, and Markets
    Video on YouTube

Miscellaneous Links and Info

  • I co-host a "faculty + friends" social group in ann arbor called Faculty Sanity.
  • A photo taken during a juggling show at MIT. Don't ask why they called me Sinbad.
  • A reporter caught me juggling on a ladder.
  • In a previous life, I used to do comedy juggling shows. I managed to win Funniest Student at UMass while studying there, and later got the chance to open for the comedian Sinbad and later Dave Chappelle. The Daily Hampshire Gazette ran a flattering article about my show right before the Sinbad appearance.
  • I love playing Racquetball.
  • I rode a bicycle from Boston to San Francisco during the summer of 2002 with two friends, Keith Vanderlinde and Damien Burke. Keith had the good sense to document the trip.
  • My mom, an expert on Anthrax and Anthrax Vaccine, has a very popular informational Anthrax Vaccine Web site and blog.
  • I used to help manage MEET: Middle East Education through Technology, an educational initiative aimed at bringing together Israeli and Palestinian youth. This project has brought me to Jerusalem three times.