![Loading Events](https://www.ri.cmu.edu/app/plugins/the-events-calendar/src/resources/images/tribe-loading.gif)
MSR Speaking Qualifier
June
![](https://www.ri.cmu.edu/app/uploads/2017/01/gandhi_dhiraj-prakashihand_2051017.jpg)
Carnegie Mellon University
3:00 pm to 4:30 pm
NSH 4305
Title: Self-Supervised Robot Learning
Abstract:
Supervised learning has been used in robotics to solve various tasks like navigation, fine manipulation, etc. While it has shown a promising result, in most cases the supervision comes from the human agent. However, relying on human is a huge bottleneck to scale up these approaches. In this thesis, we try to circumvent human supervision and try to solve the task in a self-supervised manner. More specifically, we have applied self-supervised learning to 2 tasks, for drone navigation and exploration in reinforcement learning. In drone navigation, drone first randomly collides with lots of objects and based on it, it learns a policy to avoid them. In reinforcement learning, we try to learn exploration policy in a self-supervised manner, for both stochastic and deterministic environment.
Committee:
Abhinav Gupta (advisor)
Martial Hebert
Lerrel Pinto