Samuel Dooley

sdooley1 [at] cs [dot] umd [dot] edu

[Google Scholar] [Twitter]

I am a fourth year graduate student at the University of Maryland, motivated by work that improves, studies, or changes how technology is used in applications with social impact. Together with my advisor, John P. Dickerson, I research existing technology and develop novel machine learning techniques which help people better use assistive technology and minimize technological harm. With existing technology, I've been auditing biases in academic and commercial facial recognition and facial detection systems. I also work with Elissa Redmiles to study exogenous changes to privacy attitudes and behaviors through longitudinal survey panels and in situ ad responsiveness. My work in novel machine learning advancements centers on developing mechanisms which provide machine assistance to over-burdened workers while quantifying and grappling how these new technologies alter previous human behavior (both positively and negatively). With Philip Resnik and an interdisciplinary team, I am applying these theoretical advances to mental health risk assessment.

I'll be interning wtih Amazon for Summer 2022.

I was a Visiting Scholar at MPI-SWS working with Elissa Redmiles for Summer 2021.

I have a Master's in Statistics from George Washington University, and a Bachelor's in Mathematics from University of Chicago.

Published

Angelica Goetzen, Samuel Dooley, and Elissa M Redmiles. Ctrl-Shift: How Privacy Sentiment Changed from 2019 to 2021. To appear at the 22nd Privacy Enhancing Technologies Symposium (PETS), 2022. [Link]

Marina Knittel, Samuel Dooley, and John P Dickerson. The binary atliate matching problem: approval-based matching with applicant-employer relations. To appear at the 31st International Joint Conference on Artificial Intelligence and the 25th European Conference on Artificial Intelligence (IJCAI), 2022. [Link]

Samuel Dooley, Dana Turjeman, John P Dickerson, and Elissa M Redmiles. Field Evidence of the Effects of Pro-sociality and Transparency on COVID-19 App Attractiveness. The 2022 ACM Conference on Human Factors in Computing Systems (CHI), 2022. [Link] 🎉 Best Paper Award Honorable Mention 🎉

Neehar Peri, Michael J Curry, Samuel Dooley, and John P Dickerson. Preferencenet: Encodinghuman preferences in auction design with deep learning. Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS), 2021. [Link.]

Vedant Nanda*, Samuel Dooley*, Sahil Singla, Soheil Feizi, and John P Dickerson. Fairness through robustness: Investigating robustness disparity in deep learning. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT), pages 466 - 477,2021. [Link.]

Samuel Dooley, Candice Schumann, Han-Chin Shing, John P Dickerson, and Philip Resnik. Sequential decision making in resource constrained global health settings. ML For Global Health at ICML, 2020.[Link]

Duncan C McElfresh, Samuel Dooley, Yuan Cui, Kendra Griesman, Weiqin Wang, Tyler Will, Neil Sehgal, and John P Dickerson. Can an algorithm be my healthcare proxy? Workshop on Health Intelligence at AAAI, 2020. [Link.]

Samuel Dooley, Michael Rosenberg, Elliott Sloate, Sungbok Shin, and Michelle Mazurek. Libraries' approaches to the security of public computers. 5th Workshop on Inclusive Privacy and Security at SOUPS-20, 2020. [Link.]

Darius Lam, Richard Kuzma, Kevin McGee, Samuel Dooley, Michael Laielli, Matthew Klaric, Yaroslav Bulatov, and Brendan McCord. xView: Objects in context in overhead imagery. ML for the Developing World at NeurIPS, 2018. [Link] [Press]

Eliza Mace, Keith Manville, Monica Barbu-McInnis, Michael Laielli, Matthew Klaric, and Samuel Dooley. Overhead detection: Beyond 8-bits and RGB. Naval Applications of Machine Learning, NAML 2018. [Link.]

Work Under Review

Samuel Dooley, Tom Goldstein, and John P Dickerson. Robustness Disparities in Commercial Face Detection. Under Submission. [Link] [Press]

Valeriia Cherepanova, Steven Reich, Samuel Dooley, Hossein Souri, Micah Goldblum, Tom Goldstein. A Deep Dive into Dataset Imbalance and Bias in Face Identification. Under Submission. [Link]

Samuel Dooley, Ryan Downing, George Wei, Nathan Shankar, Bradon Thymes, Gudrun Thorkelsdottir, Tiye Kurtz-Miott, Rachel Mattson, Olufemi Obiwumi, Valeriia Cherepanova, Micah Goldblum, John P Dickerson, Tom Goldstein. Comparing Human and Machine Bias in Face Recognition. Under Submission. [Link]

Samuel Dooley, Candice Schumann, Han-Chin Shing, John P Dickerson, and Philip Resnik. A multi-stage human-machine framework for mental health risk assessment. Under Submission.

Kevin Kuo, Anthony Ostuni, Elizabeth Horishny, Michael J Curry, Samuel Dooley, Ping-yeh Chiang, Tom Goldstein, and John P Dickerson. Proportionnet: Balancing fairness and revenue for auction design with deep learning. Under Submission. [Link.]

Samuel Dooley and John P Dickerson. Global best arm identification in contextual bandits. Working Paper.

Samuel Dooley and John P Dickerson. The Affiliate Matching Problem: On Labor Markets where Firms are Also Interested in the Placement of Previous Workers. Working Paper. [Link]