Seth Viren Neel 
Hi! I’m in my fifth and final year of a Ph.D. in Statistics at The Wharton School, University of Pennsylvania where I’m under the supervision of Michael Kearns and Aaron Roth. I’m supported by a 2017 NSF Graduate Fellowship . My primary research (Google Scholar) is on algorithmic fairness and differential privacy. More generally, I am interested in all mathematical aspects of machine learning and artificial intelligence. I like to design algorithms I can turn into code, some of which can be found here .
I graduated with a math degree from Harvard in 2015. Outside of research I am a co-founder of Welligence, a fintech company focused on the energy sector and backed by Quantum Energy Partners. I was named to the Forbes 30 under 30 list in the Energy category in 2019.
Research – Differential Privacy | Fairness in Machine Learning
All author ordering is strictly alphabetical. My research has two threads:
- Fairness in Machine Learning (design of new algorithms and fairness notions, in the online, bandit, batch settings)
- The study of fundamental problems in differential privacy, with an emphasis on private learning and adaptive data analysis
Preprints
Publications
A New Analysis of Differential Privacy’s Generalization Guarantees [ITCS ’20] regular talk slot!
The Role of Interactivity in Local Differential Privacy [FOCS ’19]
How to use Heuristics for Differential Privacy [FOCS ’19]
Differentially Private Objective Perturbation: Beyond Smoothness and Convexity [NEURIPS Workshop on Privacy in ML ’19]
Optimal, Truthful, and Private Securities Lending [NEURIPS Workshop on Robust AI in Financial Services ’19] selected for oral presentation!
An Empirical Study of Rich Subgroup Fairness for Machine Learning [ACM FAT* ’19, ML track]
- Code on my Github and developed further by the AlgoWatch Team
Fair Algorithms for Learning in Allocation Problems [ACM FAT* ’19, ML track]
Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness [ICML ’18, EC MD4SG ’18]
Mitigating Bias in Adaptive Data Gathering via Differential Privacy [ICML ’18]
Accuracy First: Selecting a Differential Privacy Level for Accuracy Constrained ERM [NIPS ’17, Journal of Privacy and Confidentiality ’19]
A Framework for Meritocratic Fairness of Online Linear Models [AAAI/AIES ’18]
Fair Algorithms for Infinite and Contextual Bandits [FATML ’17]
Better Fair Algorithms for Contextual Bandits [FATML ’17]
Rawlsian Fairness for Machine Learning [FATML ’16]
A Convex Framework for Fair Regression [FATML ’17]
Math stuff from College & High School
Mahalanobis Matching and Equal Percent Bias Reduction[Senior Thesis, Harvard ’15]
Plane Partitions and Domino Tilings [Intel Science Talent Search Semifinalist, ’11]
Talks & Service
- Columbia Computer Science Dept. Seminar, 11.22.19. NYC.
- NYU Tandon Computer Science Dept. Seminar, 11.14.19. NYC.
- FOCS 19, Contributed Talk. 11.11.19, Baltimore, Maryland.
- Stanford Computer Science Dept. Seminar, 3.15.19. Stanford, CA.
- FAT* 2019. Atlanta, Georgia. 1.29.18. Program Committee.
- Temple CIS Dept. Seminar, Invited Speaker. 11.29.18. Philadelphia, PA.
- Boston University Computer Science Dept. Seminar. 11.19.18. Boston, MA.
- Northeastern Computer Science Dept. Seminar. 11.8.18. Boston, MA.
- MIT CSAIL Algorithms and Complexity Seminar, 11.7.18. Cambridge, MA.
- Penn Research in Machine Learning, seminar series. 9.29.18. Slides.
- Program Committee ACM Conference on Fairness, Accountability, and Transparency 2019 (ACM FAT*)
- ICML 2018, Contributed Talk. 7.11.18, Stockholm, Sweden.
- Preventing Fairness Gerrymandering
- Mitigating Bias in Adaptive Data Gathering via Differential Privacy
- Mechanism Design for Social Good Workshop at EC 2018, Contributed Talk. 6.22.18, Ithaca, NY. Slides
- BIRS Mathematical Foundations of Data Privacy Workshop, Invited Talks. 4.29.18, Banff, Alberta.
- AAAI Workshop on Artificial Intelligence Ethics & Society. Poster Presentation. 2.3.18, New Orleans, LA.
- Neural Information Processing Systems, 12.8.17, Poster Presentation. Long Beach, CA.
- Facebook Privacy Ethics Workshop, 11.2.17, NYC.
- Theory and Practice of Differential Privacy, Contributed Talk. 10.30.17, Dallas, TX. Slides
- Fairness Accountability and Transparency in Machine Learning, Contributed Talk. New York University. 11.1.16. Video
Media
Forbes 30 under 30 in Energy 2019 The Future of Entrepreneurship is Students (Forbes, May 2018)Facebook's Racially Targeted Ads Aren't as Racist As You Think (Wired, Nov. 2016) The Best Time to Play Powerball Could Have Been 600m Ago (CNBC, Jan. 2016)
Contact
Office(s): 416B‐1 B-Wing, 3401 Walnut Street, Philadelphia, PA (CompSci), 1635 Market St. 16th Floor, Philadelphia, PA (Welligence)
Email(s): firstname.lastname@welligence.com (company) fullname@wharton.upenn.edu (academic)
