Biography

I am PhD candidate at Helmholtz AI in Munich, advised by Prof. Niki Kilbertus. I have two primary themes of interest. (1) Questions at the intersection of causality and machine learning. In particular how we can make meaningful causal inferences without making prohibitively strong assumptions, and how we can apply it to the real world. (2) I am passionate about questions related to the societal implications of AI, and particularly how we can have better controls and checks to take the good and avoid the bad that comes with rapid progress in the field.

Interests

  • AI Ethics and Fairness
  • Causality and Machine Learning
  • Partial Identification
  • Social Implications of AI

Education

  • PhD candidate (ongoing)

    Helmholtz AI, Münich

  • M.Sc. in Computer Science, 2020

    Swiss Federal Institute of Technology Lausanne (EPFL)

  • B.S. in Mathematics, 2016

    Indian Institute of Technology Kanpur (IIT Kanpur)

Publications

Stochastic causal programming for bounding treatment effects
Kirtan Padh, Jakob Zeitler, David Watson, Matt Kusner, Ricardo Silva, Niki Kilbertus
Partial identification without distributional assumptions
Kirtan Padh, Jakob Zeitler, David Watson, Matt Kusner, Ricardo Silva, Niki Kilbertus
Addressing fairness in classification with a model-agnostic multi-objective algorithm
Kirtan Padh, Diego Antognini, Emma Lejal-Glaude, Boi Faltings, Claudiu Musat
Regulatory mechanisms and algorithms towards trust in AI/ML
Eva Thelisson, Kirtan Padh, L Elisa Celis

Experience

Work and research experience

 
 
 
 
 

Doctoral Researcher

Helmholtz AI

Mar 2021 – Present Münich, Germany
Working at the intersection of causality and machine learning, as well as on the societal implications of AI.
 
 
 
 
 

Technical Advisor

AI Transparency Institute

Nov 2020 – Feb 2021 Lausanne, Switzerland
Helping with operationalizing AI regulation.
 
 
 
 
 

Research Intern

Swisscom Digital Lab

Feb 2020 – Sep 2020 Lausanne, Switzerland
Collaboratively implemented a model-agnostic multi-objective framework in PyTorch officially open-sourced through Swisscom. Closely involved in the formulation and implementation of the Ethics in AI principles of Swisscom, involving discussion with business stakeholders and members of technical teams across the company.
 
 
 
 
 

Research Assistant

Prof. Ola Svensson, EPFL

Sep 2016 – Sep 2018 Lausanne, Switzerland
Worked on finding approximation algorithms and lower bounds of various combinatorial optimization problems over graphs including local max-cut, exact matching, and the weighted tree augmentation problem.
 
 
 
 
 

Summer Strategy Analyst

Goldman Sachs

May 2016 – Jul 2016 Bengaluru, India
Designed a model for attributing profit and loss from the stock loan business to client hedge funds. Collaborated with team members to finalize the model and push to production.