thinking...solving...communicating

Satyapriya Krishna

AI Researcher

Hi! I am Satya, and I recently completed my PhD at Harvard School of Engineering and Applied Sciences (SEAS) with my thesis: "From Understanding to Improving Artificial Intelligence: New Frontiers in Machine Learning Explanations." My research focused on the trustworthy aspects of generative models, advised by Prof. Hima Lakkaraju and Prof. Finale Doshi-Velez. I also collaborated with Prof. Sameer Singh from UC Irvine and Prof. Steven Wu from Carnegie Mellon.
Previously, I worked at Google DeepMind and Meta FAIR, along with leading several research initiatives at Amazon for multiple products, including Amazon Web Services (AWS), Alexa, and Amazon Search (A9). I completed my masters from the School of Computer Science (SCS) @ Carnegie Mellon University, Pittsburgh Campus, with my research focus on building conversational agents using reinforcement learning. Please find my CV here.

avatar
News
[Jan, 2025]
Thesis: "From Understanding to Improving Artificial Intelligence: New Frontiers in Machine Learning Explanations"
Spring 2025
[Aug, 2022]
Teaching Fellow for COMPSCI 236R: Topics at the Interface between Computer Science and Economics
Fall 2022
[May, 2022]
This article highlights some of my research works contributing to Amazon's Trustworthy ML Initiatives
[April, 2022]
The Disagreement Problem was presented at TRAIT Workshop at CHI-2022
[April, 2022]
Two papers on algorithmic fairness accepted at ACL-2022
[Feb, 2022]
The Disagreement Problem was highlighted by Fortune Magazine in multiple articles
[Feb, 2022]
Received support from Nicole A. Chen and Karina A. Chen Graduate Student Research Fellowship
[Feb, 2022]
The Disagreement Problem was covered by Analytics India Magazine
[Feb, 2022]
Published our work on the disagreement problem in explainable machine learning
[Aug, 2021]
Joined Harvard SEAS as CS PhD Student
[Jun, 2021]
Our work on "Does Robustness Improve Fairness?" accepted at ACL-2021 Findings
[Feb, 2021]
VentureBeat.com coverage of our BOLD paper.
[Jan, 2021]
Our differentially private text transformation method published at EACL-2021
[Dec, 2020]
We presented a new fairness benchmark for open language generation in FAccT '21

Google Scholar

My publications

Satyapriya Krishna

LinkedIn Profile

Academic Services

Organisational Contributions

  • Co-President - Harvard SEAS Graduate Council 2022

Address

School of Engineering
Harvard University
Boston, MA

Mail Me!

Email Address

Satyapriya Krishna

LinkedIn Profile

Copyright © 2023 by Satyapriya Krishna. All rights reserved.