A picture of Karan

Karan Goel

About me. I'm a first-year Computer Science PhD student at Stanford, advised by Prof. Chris Re. At Stanford, I have collaborated with, and continue to work with Prof. Emma Brunskill, Prof. Fei-Fei Li and Prof. Jure Leskovec. I am affiliated with the Stanford AI Lab, Stanford Infolab, Stanford Human-Centered AI Institute and Statistical Machine Learning group, and have ongoing research collaborations with the Stanford Vision Lab, Stanford Medicine and Stanford Children's Health.

In Summer 2019, I will be headed to Salesforce Einstein AI Research in Palo Alto as a Deep Learning Research Intern.

Research interests. As humans, our ability to translate intent into behavior by leveraging our knowledge is amazing! My goal is to build autonomous agents that can do the same. Can we design agents that help us achieve our goals, by learning through interaction with the world and representing and utilizing their prior knowledge?

I'm especially interested in building agents that can learn by observing human behavior, and advancing the theory and practice for these problems. I think about these questions mainly through the lens of Reinforcement Learning, but I also draw from other areas of Statistical Machine Learning. Another area I'm passionate about is using AI to solve real problems, by creating systems for AI-assisted education and healthcare.

Historical note. I graduated with a Master's degree from the Department of Machine Learning at Carnegie Mellon University in 2018, where I was a recepient of the Siebel Foundation's 2017-18 Scholarship. Concurrently, I spent time at Stanford (2017-2018) as a visiting Research Assistant in the AI Lab with Prof. Emma Brunskill. I did my undergrad at the Indian Institute of Technology - Delhi (IITD) where I spent 5 amazing years majoring in EECS. At IITD, I got into AI research while working with Prof. Mausam. I also greatly enjoyed collaborating with Prof. Aditya Parameswaran and Prof. Hari Sundaram during an internship at the University of Illinois at Urbana-Champaign.

If you'd like to collaborate on research, chat with me about AI, network with me or get free advice then drop me a line.

Current affiliations


News Publications

  • [May '19] I enjoyed attending ICLR and presenting our work!
  • [Mar '19] The first grant I've ever applied for was funded! With Peyton (MVP), Dan and Sumit, we'll try to understand whether video can monitor respiratory conditions and improve turnaround time in the pediatrics ER. $60k thanks to the Stanford HAI seed grants!
  • [Mar '19] My next rotation (in Spring quarter) will be with Jure!
  • [Feb '19] I'll be headed to Salesforce Einstein AI Research this summer as a Deep Learning Research Intern. Very excited to work with Caiming and Richard!
  • [Jan '19] It's time for Winter quarter and my rotation with Chris!
  • [Jan '19] Our paper on "PLOTS: Procedure Learning from Observations using Subtask Structure" was accepted to AAMAS 2019! Led by Tong and jointly with Emma.
  • [Dec '18] My paper with Emma on "Learning Procedural Abstractions and Evaluating Discrete Latent Temporal Structure" was accepted to ICLR 2019!
  • [Dec '18] I attended NeurIPS 2018 in Montreal (beautiful city)! I presented a poster on "Learning Procedural Abstractions" at the Infer2Control workshop as well as a fun live demonstration(!!) on our automated piano tutoring system for learning from video called "Demo2Tutor" with Tong (joint with Jonathan and Emma).
  • [Sep '18] I joined Stanford as a CS PhD student! Rotating with Emma and Fei-Fei in Autumn quarter!
  • [May '18] Graduated with a Master's in Machine Learning from CMU! CMU was a great experience (Larry's stats courses are awesome) and I'll miss it!
  • [March '18] I've decided to take Stanford's offer and I'll be joining the Stanford CS PhD program this fall!
Conference
Workshop
Preprint
Demonstration
PLOTS: Procedure Learning from Observations using Subtask Structure Tong Mu, Karan Goel and Emma Brunskill AAMAS 2019
Automatic Curriculum Generation Applied to Teaching Novices a Short Bach Piano Segment Tong Mu, Karan Goel, Jonathan Bragg and Emma Brunskill Live Demonstration
NeurIPS 2018
Learning Procedural Abstractions Karan Goel and Emma Brunskill Infer2Control Workshop
NeurIPS 2018
Shared Autonomy for Interactive Systems Sharon Zhou, Tong Mu, Karan Goel, Michael Bernstein and Emma Brunskill Poster
UIST 2018
Program2Tutor: Combining Automatic Curriculum Generation with Multi-Armed Bandits for Intelligent Tutoring Systems Tong Mu, Karan Goel and Emma Brunskill Workshop on Teaching Machines, Humans and Robots
NeurIPS 2017
Optimal Hierarchical Policy Extraction from Noisy Imperfect Demonstrations Karan Goel, Tong Mu and Emma Brunskill Hierarchical Reinforcement Learning Workshop & Deep Reinforcement Learning Symposium
NeurIPS 2017
Importance-Sampled Option Critic for More Sample-Efficient Reinforcement Learning Karan Goel and Emma Brunskill Hierarchical Reinforcement Learning Workshop & Deep Reinforcement Learning Symposium
NeurIPS 2017
It's just a matter of perspectives: Crowd Powered Consensus Organization of Corpora Ayush Jain, Karan Goel, Joon Young-Seo, Andrew Kuznetsov, Aditya Parameswaran and Hari Sundaram Arxiv
Sample Efficient Policy Search for Optimal Stopping Domains Karan Goel, Christoph Dann and Emma Brunskill RLDM 2017 & IJCAI 2017
POMDP-Based Worker Pool Selection for Crowdsourcing Shreya Rajpal, Karan Goel and Mausam CrowdML Workshop
ICML 2015