|
Jason AltschulerAssistant Professor of Statistics and Data Science, UPenn
Office: ARB 315 |
I am creating/organizing the UPenn Optimization Seminar. Join us Thursdays 12-1pm in SHDH 1201.
I am an Assistant Professor at UPenn in the Wharton Department of Statistics and Data Science, the Department of Electrical Engineering (by courtesy), and the Department of Computer Science (by courtesy). I am also an affiliated faculty member in Applied Mathematics, the PRiML Center for Machine Learning, and the Warren Center for Networks and Data Science.
My research interests are broadly at the interface of optimization, probability, and machine learning, with a focus on the design and analysis of large-scale algorithms.
Previously, I received my undergrad degree from Princeton, and my PhD in Electrical Engineering and Computer Science from MIT, where I recieved the Sprowls Award for the best MIT thesis in AI & Decision Making. I am interested in practical implementations as well as theory, and have spent time in both tech (Apple Research, Google Research) and Wall Street (DE Shaw, Tower Research).
In my free time, I like to ski, play tennis, and eat too many cookies. I also like chess and, in a past life, I obtained an International Master norm in Spain. In another past life I swam on the Oxford varsity team.
Shifted Composition II: Shift Harnack Inequalities and Curvature Upper Bounds
Jason Altschuler, Sinho Chewi.
Resolving the mixing time of the Langevin Algorithm to its stationary distribution for log-concave sampling
Jason Altschuler, Kunal Talwar.
Flows, Scaling, and Entropy Revisited: a Unified Perspective via Optimizing Joint Distributions
Jason Altschuler
Shifted Composition I: Harnack and Reverse Transport Inequalities
Jason Altschuler, Sinho Chewi.
Acceleration by Stepsize Hedging II: Silver Stepsize Schedule for Smooth Convex Optimization
Jason Altschuler, Pablo Parrilo.
Acceleration by Stepsize Hedging I: Multi-Step Descent and the Silver Stepsize Schedule
Jason Altschuler, Pablo Parrilo.
Faster high-accuracy log-concave sampling via algorithmic warm starts
Jason Altschuler, Sinho Chewi.
On the Privacy of Noisy Stochastic Gradient Descent for Convex Optimization
Jason Altschuler, Jinho Bok, Kunal Talwar.
Kernel approximation on algebraic varieties
Jason Altschuler, Pablo Parrilo.
Polynomial-time algorithms for Multimarginal Optimal Transport problems with structure
Jason Altschuler, Enric Boix-Adsera.
Asymptotics for semi-discrete entropic optimal transport
Jason Altschuler, Jonathan Niles-Weed, Austin Stromme.
Near-linear convergence of the Random Osborne algorithm for Matrix Balancing
Jason Altschuler, Pablo Parrilo.
Approximating Min-Mean-Cycle for low-diameter graphs in near-optimal time and memory
Jason Altschuler, Pablo Parrilo.
Wasserstein barycenters are NP-hard to compute
Jason Altschuler, Enric Boix-Adsera.
Wasserstein barycenters can be computed in polynomial time in fixed dimension
[poster, slides]
Jason Altschuler, Enric Boix-Adsera.
Hardness results for Multimarginal Optimal Transport problems
Jason Altschuler, Enric Boix-Adsera.
Online learning over a finite action set with limited switching
[poster, talk]
Jason Altschuler, Kunal Talwar.
Lyapunov Exponent of Rank One Matrices: Ergodic Formula and Inapproximability of the Optimal Distribution
[slides]
Jason Altschuler, Pablo Parrilo.
Best arm identification for contaminated bandits
Jason Altschuler, Victor-Emmanuel Brunel, Alan Malek.
Inclusion of forbidden minors in random representable matroids
Jason Altschuler, Elizabeth Yang.
Rapid analysis and exploration of fluorescence microscopy images
Benjamin Pavie, Satwik Rajaram, Austin Ouyang, Jason Altschuler, Robert Steininger, Lani Wu, Steven Altschuler.
Shifted Interpolation for Differential Privacy
Jinho Bok, Weijie Su, Jason Altschuler.
Faster high-accuracy log-concave sampling via algorithmic warm starts
Jason Altschuler, Sinho Chewi.
Privacy of Noisy Stochastic Gradient Descent: more iterations without more privacy loss
Jason Altschuler, Kunal Talwar.
Averaging on the Bures-Wasserstein manifold: dimension-free convergence of gradient descent
Jason Altschuler, Sinho Chewi, Patrik Gerber, Austin Stromme.
Lyapunov Exponent of Rank One Matrices: Ergodic Formula and Inapproximability of the Optimal Distribution
[slides]
Jason Altschuler, Pablo Parrilo.
Massively scalable Sinkhorn distances via the Nyström method
Jason Altschuler, Francis Bach, Alessandro Rudi, Jonathan Niles-Weed.
Online learning over a finite action set with limited switching
[poster, talk]
Jason Altschuler, Kunal Talwar.
Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration
[poster]
Jason Altschuler, Jonathan Weed, Philippe Rigollet.
Greedy column subset selection: new bounds and distributed algorithms
[poster, slides]
Jason Altschuler, Aditya Bhaskara, Gang Fu, Vahab Mirrokni, Afshin Rostamizadeh, Morteza Zadimoghaddam.
Transport and Beyond: Efficient Optimization over Probability Distributions
George M. Sprowls Award for Best MIT PhD Thesis in Artificial Intelligence & Decision Making
Greed, hedging, and acceleration in convex optimization
Minimax rates for online learning with limited decision changes (superseded by this paper in MOR/COLT)
Probabilistic linear Boolean classification