About
My name is Ayush Kanodia. I’m a senior PhD candidate in Computer Science and MA candidate in Economics at Stanford University, currently in my fifth year. I mix Machine Learning and Economics. I am lucky to be advised by Susan Athey, and to be co advised by Emma Brunskill.
Research Interests
I work at the intersection of Economics and Machine Learning, using methods from each into the other. My focus areas are machine learning applied to economic structural modeling and causal inference (and vice versa), and recommendation systems. Additionally, I like to learn about/use Deep Learning methods, Reinforcement Learning and Language Processing methods towards these goals.
Research
- Torch-Choice: A PyTorch Package for Large-Scale Choice Modelling with Python, Du, Tianyu; Kanodia, Ayush; Athey, Susan, arXiv preprint arXiv:2304.01906, 2023
- Personalized Recommendations in EdTech: Evidence from a Randomized Controlled Trial, Agrawal, Keshav; Athey, Susan; Kanodia, Ayush; Palikot, Emil, arXiv preprint arXiv:2208.13940, 2022
- Welfare effects of personalized rankings, Donnelly, Robert; Kanodia, Ayush; Morozov, Ilya, Available at SSRN 3649342, 2022
- Learning Transferrable Representations of Career Trajectories for Economic Prediction, Vafa, Keyon; Palikot, Emil; Du, Tianyu; Kanodia, Ayush; Athey, Susan; Blei, David M, arXiv preprint arXiv:2202.08370, 2022
- policytree: Policy learning via doubly robust empirical welfare maximization over trees, Sverdrup, Erik; Kanodia, Ayush; Zhou, Zhengyuan; Athey, Susan; Wager, Stefan, Journal of Open Source Software, vol. 5, no. 50, pp. 2232, 2020
- Optimal cost almost-sure reachability in POMDPs, Chatterjee, Krishnendu; Chmelik, Martin; Gupta, Raghav; Kanodia, Ayush, Artificial Intelligence, vol. 234, pp. 26-48, 2016
- Qualitative analysis of POMDPs with temporal logic specifications for robotics applications, Chatterjee, Krishnendu; Chmelik, Martin; Gupta, Raghav; Kanodia, Ayush, 2015 IEEE International Conference on Robotics and Automation (ICRA), pp. 325-330, 2015
- Biological auctions with multiple rewards, Reiter, Johannes G; Kanodia, Ayush; Gupta, Raghav; Nowak, Martin A; Chatterjee, Krishnendu, Proceedings of the Royal Society B: Biological Sciences, vol. 282, no. 1812, pp. 20151041, 2015
- Deciphering Indian inflationary expectations through text mining: an exploratory approach, Banerjee, Ashok; Kanodia, Ayush; Ray, Partha, Indian Economic Review, vol. 56, pp. 49-66, 2021
Prior experience
In reverse chronological order
- Software Engineer, Google (London), working on Text-to-Speech
- Researcher, IIM Calcutta, where I worked on applying Machine Learning to Macroeconomics and Finance. I worked on Estimating Inflation using News Articles and we built textplor.com in my time there. I was advised by Partha Ray and Ashok Banerjee.
- UG Thesis, IIT Bombay, where I worked with Soumen Chakrabarti and S Sudarshan.
- Research Intern, Microsoft Research (Redmond) Mobility and Networking Group, advised by Stefan Saroiu.
- Researcher, IST Austria (Vienna), advised by Krishnendu Chatterjee.
Past Education
- Graduation - B.Tech (Computer Science) with Honors and Minor in Applied Statistics and Informatics, IIT Bombay, India
- High School - Don Bosco School Park Circus, Kolkata, India