Avatar

Kanishk Gandhi

Stanford University

New York University

PathAI

Biography

Hello visitor, so glad that you are visiting my webpage!! Unfortunately the content on this page is outdated, and me in the picture to your left is from a time when I was visiting LA, thought corona was just a beer and had one fabulous fedora that I wore everywhere. A lot has changed since then, I've moved from New York to the Bay Area, started a PhD in CS at Stanford, and have doubled my collection of fabulous fedoras to TWO. I will update this page soon! Please feel free to send me an email at {first}.{last}@stanford.edu in the meantime.

How do children learn new concepts from a few examples? Why does an algorithm that has seen millions of minutes of videos not understand the fundamental properties of objects, but we do so easily? How can infants from arguably 15 months of age understand that others have desires and beliefs that could be different from theirs, while autonomous agents trained for more than all of human existence fail to do so? I am a research scientist at NYU working with Prof. Brenden Lake and Prof. Moira Dillon at the Human & Machine Learning Lab on understanding human cognitive abilities in an effort to advance current AI algorithms.

Interests

  • Artificial Intelligence
  • Cognitive Development

Education

  • M.S. in ECE, 2020

    New York University

  • B.Tech. in Electrical Engineering, 2018

    IIT Kanpur

Publications

December 2020 Working Paper

Benchmarks and Baselines for Intuitive Psychology

To achieve human-like thinking and performance, machine learning systems need to comprehend the autonomous agents in their environment. Inspired by work in developmental psychology, we present challenges for machines that would test their abilities to make theory of mind inferences. Similar to developmental studies, we use a violation of expectation paradigm where the machine must predict the plausibility of a video sequence. We further present a few baselines to understand the challenges that might be faced in achieving a good score on this benchmark.

Talk Slides
July 2019 NeurIPS 2020

Mutual Exclusivity as a Challenge for Deep Neural Networks

Children use the mutual exclusivity (ME) bias to learn new words, while standard neural nets show the opposite bias, hindering learning in naturalistic scenarios such as lifelong learning.

PDF New Scientist Coverage arXiv

Teaching / Mentorship

  • I have been an instructor at the Machine Learning Summer School (May 2018) at IIT Kanpur which was a 10 days school covering the basics of machine learning and deep learning. The course website is available here.
  • I have been an Academic Mentor for the course Fundamentals of Computing for the academic year (2015-16) where I mentored Freshmen in an introductory programming course.