About

I am an Assistant Research Professor at Cornell's Center for Data Science for Enterprise and Society, co-hosted by Nikhil Garg and danah boyd.

My work is grounded in the algorithmic foundations of responsible computing. I broadly study mechanisms to resolve potential harms incurred by AI-driven decision-making, with a focus on techniques that are scalable for real-world use, holistically consider an algorithm in its broader context, and flexibly incorporate human input. I am particularly interested in methods that allow efficient post-deployment model adjustment.

I received my PhD from the University of Pennsylvania, advised by Aaron Roth and Michael Kearns. Prior to Penn, I worked at Boston University as a software engineer with SAIL and with the Harvard Privacy Tools Project. I received my bachelor's degree from Reed College in Portland, OR.

Announcements

Fall Travel

I have quite a bit of upcoming travel: please reach out if you are at any of these places and want to grab coffee and/or stand at a whiteboard and munch on some fun problems together.

I will be at the University of Colorado Boulder's CS theory seminar on Friday September 18th speaking about information aggregation, and will stick around the next week if you want to meet and chat!
Natalie Collina and I are giving a modified version of our calibration and collaboration tutorial at CMU's Human-AI Complimentarity for Decision Making workshop on dynamic alignment.
I will be at the University of Waterloo hosted by Shai Ben-David at the beginning of October, and will be giving the Algorithms and Complexity seminar on September 30th.
Then, for a bit of a vibe change, I will be at 4S in Toronto, speaking on my work on human-AI collaboration and recourse from the lens of sociotechnical system design. This is a project I have been working on with my co-host at Cornell, danah boyd, whose book you should totally go pre-order!
In November, I'll be at Informs before heading to Australia to give a talk at the University of Sydney in mid-November, and will be in the area through Neurips.

Recent News

August 2026 I had a wonderful time visiting Berkeley for the Public Feedback for AI Evaluations workshop and microconference, run by the fantastic Jessica Dai with the moral and logistics support of Ben Recht. I spoke about my ideas for bias bounties and incorporation of human feedback into model design, and thought Jess did a fabulous job building a great community. You can read part of her retrospective about the conference here. A huge thank you to the Survival and Flourishing fund for making the event possible.

July 2026 It was an absolute pleasure to meet with Chenglin Fan's students at Seoul National University's Algorithmic Foundations of Data Science lab. They are a great group of bright and articulate folks, and I look forward to following their research careers. July 2026 Co-organized the ICML 2026 tutorial Calibration, Decisions, and Collaboration in Learning with Aaron Roth, Natalie Collina. The tutorial website now includes our full slide deck and a recording of the tutorial. It is, we like to think, accessible to a much broader audience than your standard ML theory talk, with an emphasis on intuition-building, while still being of use to those of you who really want to see slides full of math.

June 2026

My and Nikhil Garg's whitepaper on meaningful model multiplicity was accepted at the ICML 2026 AI4Good Workshop! See you Friday July 10th in Grand Ballroom 103. Co-organized the Learning Theory Alliance's COLT 2026 mentorship event with Gautam Kamath, Bingbin Lin, and Jess Sorrell. Invited to the University of Zurich to speak at the DM3L's Rethinking Intelligence: Mathematics, Models, and Responsibility symposium. Thrilled to announce that I am now affiliated with the Structural Democracy Fellows. Stay tuned for some exciting upcoming projects on democracy, information aggregation, and algorithmic nudges. My and danah boyd's WIP has been accepted to 4S 2026. See you in October!

Publications

SODA 2026

Collaborative Prediction: Tractable Information Aggregation via Agreement

Natalie Collina, Ira Globus-Harris, Surbhi Goel, Varun Gupta, Aaron Roth, Mirah Shi

FORC 2025

Model Ensembing for Constrained Optimization

Ira Globus-Harris, Varun Gupta, Michael Kearns, Aaron Roth

FAccT 2024

Diversified Ensembling: An Experiment in Crowdsourced Machine Learning

Ira Globus-Harris, Declan Harrison, Michael Kearns, Pietro Perona, Aaron Roth

ICML 2023

Multicalibration as Boosting for Regression

Ira Globus-Harris, Declan Harrison, Michael Kearns, Aaron Roth, Jessica Sorrell

Oral Presentation

AIES 2023

Multicalibrated Regression for Downstream Fairness

Ira Globus-Harris, Varun Gupta, Christopher Jung, Michael Kearns, Jamie Morgenstern, Aaron Roth

FAccT 2022

An Algorithmic Framework for Bias Bounties

Ira Globus-Harris, Michael Kearns, Aaron Roth

Journal of Survey Statistics and Methodology 2022

Non-parametric Differentially Private Confidence Intervals for the Median

Joerg Drechsler, Ira Globus-Harris, Audra McMillan, Jayshree Sarathy, Adam Smith

FORC 2021

Lexicographically Fair Learning: Algorithms and Generalization

Emily Diana, Wesley Gill, Ira Globus-Harris, Michael Kearns, Aaron Roth, Saeed Sharifi-Malvajerdi

PETS 2019

Improved Differentially Private Analysis of Variance

Marika Swanberg, Ira Globus-Harris, Iris Griffith, Anna Ritz, Adam Groce, Andrew Bray

SOUPS 2019

From Usability to Secure Computing and Back Again

Lucy Qin, Andrei Lapets, Frederick Jansen, Peter Flockhart, Kinan Dak Albab, Ira Globus-Harris

IEEE SecDev 2019

Tutorial: Deploying Secure Multi-Party Computation on the Web Using JIFF

Kinan Dak Albab, Rawane Issa, Andrei Lapets, Peter Flockhart, Lucy Qin, Ira Globus-Harris

Awards & Recognitions

2024 AWS AI ASSET Fellow

2023 EECS Rising Star, Georgia Tech

(Selected) Invited Talks

June 2026

Algorithmic Hooks for Collaborative and Fair Machine Learning

University of Zurich

April 2026

Collaborative Machine Learning

Cornell Theory Seminar

Feb 2026