Justin Hsu

Contacts

UPenn
GRW 561
Levine Hall

Email
email@justinh.su

About Me

I am a graduate student in the Department of Computer Science at the University of Pennsylvania. There, I am associated with the Theory Group, the PL Club, and the Privacy Group. I am very fortunate to be co-advised by Benjamin Pierce and Aaron Roth.

I am supported by the Simons Graduate Fellowship in Theoretical Computer Science. Previously, I was awarded an Honorable Mention for the NSF Graduate Research Fellowship Program.

Research Interests

I spend my time thinking about questions in verification, programming languages, and algorithms. My work has mostly centered on differential privacy, a rigorous definition of privacy that is currently under extensive study.

From the verification side, I am interested in using formal methods—such as type systems and program logics—to verify that programs are differentially private. From the algorithms side, I am interested in applications of differential privacy to optimization, machine learning, and mechanism design.

News
Bibliography

Drafts

Coupling proofs are probabilistic product programs
Gilles Barthe, Benjamin Grégoire, Justin Hsu, and Pierre-Yves Strub.
[eprint]
Formal certification of randomized algorithms
Gilles Barthe, Thomas Espitau, Marco Gaboardi, Benjamin Grégoire, Justin Hsu, and Pierre-Yves Strub.
[Paper]
A semantic account of metric preservation
Arthur Azevedo de Amorim, Ikram Cherigui, Marco Gaboardi, Justin Hsu, and Shin-ya Katsumata.
[Paper]
Computer-aided verification in mechanism design
Gilles Barthe, Marco Gaboardi, Emilio Jesús Gallego Arias, Justin Hsu, Aaron Roth, and Pierre-Yves Strub.
[eprint]

Journals and Surveys

2016
Programming language techniques for differential privacy
Gilles Barthe, Marco Gaboardi, Justin Hsu, and Benjamin C Pierce.
ACM SIGLOG News, 3(1):34–53, January 2016.
[eprint]

Conferences

2016
Advanced probabilistic couplings for differential privacy
Gilles Barthe, Noémie Fong, Marco Gaboardi, Benjamin Grégoire, Justin Hsu, and Pierre-Yves Strub.
In ACM SIGSAC Conference on Computer and Communications Security (CCS), Vienna, Austria.
[eprint]
To appear.
2016
Differentially private Bayesian programming
Gilles Barthe, Gian Pietro Farina, Marco Gaboardi, Emilio Jesús Gallego Arias, Andy Gordon, Justin Hsu, and Pierre-Yves Strub.
In ACM SIGSAC Conference on Computer and Communications Security (CCS), Vienna, Austria.
[eprint]
To appear.
2016
Synthesizing probabilistic invariants via Doob’s decomposition
Gilles Barthe, Thomas Espitau, Luis María Ferrer Fioriti, and Justin Hsu.
In International Conference on Computer Aided Verification (CAV), Toronto, Ontario.
[eprint]
2016
A program logic for union bounds
Gilles Barthe, Marco Gaboardi, Benjamin Grégoire, Justin Hsu, and Pierre-Yves Strub.
In International Colloquium on Automata, Languages and Programming (ICALP), Rome, Italy.
[eprint]
2016
Proving differential privacy via probabilistic couplings
Gilles Barthe, Marco Gaboardi, Benjamin Grégoire, Justin Hsu, and Pierre-Yves Strub.
In IEEE Symposium on Logic in Computer Science (LICS), New York, New York.
[eprint] [Slides]
2016
Do prices coordinate markets?
Justin Hsu, Jamie Morgenstern, Ryan Rogers, Aaron Roth, and Rakesh Vohra.
In ACM SIGACT Symposium on Theory of Computing (STOC), Cambridge, Massachusetts.
[eprint]
2016
Jointly private convex programming
Justin Hsu, Zhiyi Huang, Aaron Roth, and Zhiwei Steven Wu.
In ACM–SIAM Symposium on Discrete Algorithms (SODA), Arlington, Virginia.
[eprint] [Slides]
2015
Relational reasoning via probabilistic coupling
Gilles Barthe, Thomas Espitau, Benjamin Grégoire, Justin Hsu, Léo Stefanesco, and Pierre-Yves Strub.
In International Conference on Logic for Programming, Artificial Intelligence and Reasoning (LPAR), Suva, Fiji.
[eprint] [Slides]
2015
Online assignment with heterogeneous tasks in crowdsourcing markets
Sepehr Assadi, Justin Hsu, and Shahin Jabbari.
In Conference on Human Computation & Crowdsourcing (HCOMP), San Diego, California.
[eprint]
2015
Death, taxes, and formal verification (Abstract)
Justin Hsu.
In Summit on Advances in Programming Langugaes (SNAPL), Asilomar, California.
[Paper] [Slides]
2015
A Theory AB toolbox
Marco Gaboardi and Justin Hsu.
In Summit on Advances in Programming Langugaes (SNAPL), Asilomar, California.
[Paper] [Slides]
2015
Higher-order approximate relational refinement types for mechanism design and differential privacy
Gilles Barthe, Marco Gaboardi, Emilio Jesús Gallego Arias, Justin Hsu, Aaron Roth, and Pierre-Yves Strub.
In ACM SIGPLAN–SIGACT Symposium on Principles of Programming Languages (POPL), Mumbai, India.
[eprint] [Slides]
2014
Really naturally linear indexed type-checking
Arthur Azevedo de Amorim, Marco Gaboardi, Emilio Jesús Gallego Arias, and Justin Hsu.
In Symposium on Implementation and Application of Functional Programming Languages (IFL), Boston, Massachusetts.
[eprint] [Slides]
2014
Proving differential privacy in Hoare logic
Gilles Barthe, Marco Gaboardi, Emilio Jesús Gallego Arias, Justin Hsu, César Kunz, and Pierre-Yves Strub.
In IEEE Computer Security Foundations Symposium (CSF), Vienna, Austria.
[eprint]
2014
Differential privacy: An economic method for choosing epsilon
Justin Hsu, Marco Gaboardi, Andreas Haeberlen, Sanjeev Khanna, Arjun Narayan, Benjamin C Pierce, and Aaron Roth.
In IEEE Computer Security Foundations Symposium (CSF), Vienna, Austria.
[eprint] [Slides]
2014
Privately solving linear programs
Justin Hsu, Aaron Roth, Tim Roughgarden, and Jonathan Ullman.
In International Colloquium on Automata, Languages and Programming (ICALP), Copenhagen, Denmark.
[eprint] [Slides]
2014
Dual Query: Practical private query release for high dimensional data
Marco Gaboardi, Emilio Jesús Gallego Arias, Justin Hsu, Aaron Roth, and Zhiwei Steven Wu.
In International Conference on Machine Learning (ICML), Beijing, China.
[eprint] [Slides] [Poster]
2014
Private matchings and allocations
Justin Hsu, Zhiyi Huang, Aaron Roth, Tim Roughgarden, and Zhiwei Steven Wu.
In ACM SIGACT Symposium on Theory of Computing (STOC), New York, New York.
[eprint] [Poster]
2013
Towards dependently typed Haskell: System FC with kind equality
Stephanie Weirich, Justin Hsu, and Richard A Eisenberg.
In ACM SIGPLAN International Conference on Functional Programming (ICFP), Boston, Massachusetts.
[Paper]
2013
Differential privacy for the analyst via private equilibrium computation
Justin Hsu, Aaron Roth, and Jonathon Ullman.
In ACM SIGACT Symposium on Theory of Computing (STOC), Palo Alto, California.
[eprint] [Slides]
2013
Automatic sensitivity analysis using linear dependent types
Marco Gaboardi, Emilio Jesús Gallego Arias, Andreas Haeberlen, Justin Hsu, and Benjamin C Pierce.
In International Workshop on Foundational and Practical Aspects of Resource Analysis (FOPARA), informal proceedings.

2013
Linear dependent types for differential privacy
Marco Gaboardi, Andreas Haeberlen, Justin Hsu, Arjun Narayan, and Benjamin C Pierce.
In ACM SIGPLAN–SIGACT Symposium on Principles of Programming Languages (POPL), Rome, Italy.
[Paper]
2012
Distributed private heavy hitters
Justin Hsu, Sanjeev Khanna, and Aaron Roth.
In International Colloquium on Automata, Languages and Programming (ICALP), Warwick, England.
[eprint] [Slides]
Thanks to Raef Bassily and Adam Smith for spotting an error, now fixed.