Avatar

Amit Sharma

Principal Researcher

Microsoft Research India

2022.07.25: Talk on necessity of causal inference for out-of-distribution generalization in prediction and decision-making at the Technion, Israel. [Slides]

2022.05.31: DoWhy library for causal inference evolves to an independent py-why org to foster wider collaboration. Contributions welcome! [Blog][Github][Arxiv]

2021.12.02: Talk on Causal Inference for Machine Learning: Generalization, Explanation and Fairness; at the UK Office for National Statistics. [Slides]

2020.12.03: Emre and I gave a Microsoft Research webinar on causal inference and its implications for machine learning. [Video]

2020.08.13: Featured on the Humans of AI podcast. [Apple Podcasts][Spotify]

2020.07.23: Session on causal machine learning with Elias Bareinboim, Susan Athey, and Cheng Zhang. [Video]

2020.06.30: MatchDG: Learning causal predictive models that generalize to new distributions [Paper][Code]

2019.05.30: DiCE: Using counterfactual examples to explain machine learning. [Paper][Python Library][Blog]

2018.08.19: Emre and I gave a tutorial on causal inference at KDD. [Slides]


Data tells stories. My research aims to tell the causal story.

As machine learning systems move into societally critical domains such as healthcare, education, finance and criminal justice, questions on their impact gain fundamental importance. The key insight in my work is to consider modern algorithms as interventions, just like a medical treatment or an economic policy. Unlike typical interventions studied in social and biomedical sciences, however, algorithmic interventions can be arbitrarily complex. I work on developing methods to estimate causal impact of such systems and build algorithms that optimize the causal effect. I am also passionate about designing new interventions for societal impact, especially in healthcare.

If you are interested in working with me at MSR India, drop me an email. We hire interns throughout the year. There are also postdoctoral positions available. Additionally, if you are an undergraduate or a masters student, our lab runs an excellent pre-doctoral Research Fellowship program.

Read more…

Interests

  • Causal Inference
  • Trustworthy Machine Learning
  • Technology and Mental Health

Education

  • PhD in Computer Science, 2015

    Cornell University

  • B.Tech. in Computer Science, 2010

    IIT Kharagpur

Recent Talks

Teaching causal reasoning to language models

Large language models (LLMs) have demonstrated remarkable accuracy in identifying cause-and-effect relationships across diverse …

Causality and Large Language Models: A new frontier

A key challenge in conducting causal analyses is that identifying the correct assumptions, such as the causal graph, needs considerable …

Causal Inference in Recommender Systems

What is the impact of a recommender system? In a typical three-way interaction between users, items and the platform, a recommender …

The impact of computing systems | Causal inference in practice

Computing and machine learning systems are affecting almost all parts of our lives and the society at large. How do we formulate and …

Measuring Effectiveness of ML Systems

Recent Posts

Trip report from ACM COMPASS: 2nd Conference on Computing and Sustainable Societies

Last year, I attended the inaugural ACM conference on Computing and Sustainable Societies (COMPASS) and was immediately sold on the …

A Gentle Introduction to Causal Inference

That we find out the cause of this effect, Or rather say, the cause of this defect, For this effect defective comes by cause. …

A Simple Guide to Doubly Robust Estimation

Two roads diverged in a wood, and I— I took the one less traveled by, But if I could go back, I will try To take both: why …

Cumulative Distribution Plots for Frequency Data in R

R has some great tools for generating and plotting cumulative distribution functions. However, they are suited for raw data, not when …

Contact