Ashutosh Saxena

Assistant Professor,
Microsoft Faculty Fellow, and Sloan Fellow.
Department of Computer Science,
Cornell University.

Education: PhD, Stanford University with Prof. Andrew Y. Ng.

321 Gates Hall, Ithaca, NY 14853.
Tel: (607)255-7017, Fax: (607)255-4428
asaxena at cs.cornell dot edu

Research Interests

Artificial Intelligence:  Machine Learning,  Robotics,  Computer Vision / Perception.

Robot Learning
Data-driven learning of new perception/ manipulation skills.
3D depths from a single still image.
3D Perception/vision
Scene/activity understanding from RGBD data.
Hallucinating Humans
Unsupervised learning of spatio-temporal affordances.

How can robots operate autonomously in unstructured environments? I design new machine learning algorithms to help robots perceive the dynamic environments so that they can perform tasks such as manipulating objects in the environment, perform assistive tasks, and more. Perceiving the environment includes 3D scene labeling, activity detection, reasoning through humans, and in general mapping the high-dimensional perceptual space to robotic actions in time. One key challenge is how to formulate traditional robotics problems of perception and planning into data-driven learning problems.

Research Group

PhD students: Yun Jiang, Hema Koppula, Ian Lenz, Jaeyong Sung, Ashesh Jain. Dipendra K Misra, Chenxia Wu.

I co-advised Congcong Li, who is now a research scientist at Google.

Other students I have worked with are: Mevlana Gemici, Wen H Lui, Rudhir Gupta, Changxi Zheng (now assistant professor at Columbia University), Paul Heran Yang, Adarsh Kowdle, Zhaoyin Jia, and Marcus Lim.

Recent News
See our workshop at RSS'14: Planning for Robots: Learning vs Humans.

Our 5th RGB-D workshop at RSS'14: Vision vs Robotics!

Our special issue on autonomous grasping and manipulation is out!

Saxena's Robot Learning Lab projects were featured in BBC World News.

Daily Beast comments about Amazon's predictive delivery and Saxena's predictive robots.

Zhaoyin Jia's paper on physics-based reasoning for RGB-D image segmentation, an oral at CVPR'13, is now conditionally accepted in IEEE TPAMI.

Vaibhav Aggarwal was awarded ELI'14 research award for his work with Ashesh Jain.

Koppula's video on reactive robotic response was the finalist for best video award at IROS 2013.

Ashesh Jain's NIPS'13 paper on learning preferences in trajectories was mentioned in Discovery Channel Daily Planet, Techcrunch, FOX News, NBC News and several others. (Watch the video!)

Saxena gave invited talks at the AI-based Robotics, at the Caging for manipulation, and at the Developmental and Social Robotics workshops at IROS 2013.

Koppula's paper on anticipation won the best student paper award at the Robotics Science and Systems (RSS) conference. It was also covered in NBCNews, LA Times, Wired magazine, FOX News, etc., and made fun of in the Daily Show (Comedy Central) by Lewis Black.

Rudhir Gupta was awarded best Masters project in 2013 at Cornell for this work on RGB-D human activity detection.

Koppula's, Jiang's and Lenz's papers accepted in RSS'13, with Koppula's paper as a full oral.

Lenz's deep learning now helping Baxter grasp new objects.

Yun Jiang's paper on 3D object detection using hallucinated humans presented as an oral at CVPR'13.

Saxena received NSF Career Award for research in robotic perception.

Hema Koppula's paper on detecting human activities from RGB-D videos accepted in IJRR, and a followup paper in ICML'13.

Saxena named one of the seven Microsoft Faculty Fellows for 2012.

Congcong Li passed her PhD defense for her work on Feedback enabled learning models and its applications.

Marcus Lim was awarded academic excellence award for almost perfect A+ GPA and for his work on arranging objects.

CUAir, an undergraduate team I advised, won second place at the AUVSI competition among 35 teams. They were placed first in the mission (flight) performance category.

Prof. Saxena won Google faculty research award, IARPA Finder, and others.

Prof. Saxena is named a co-chair of the IEEE Technical committee on robot learning.

Prof. Saxena is serving as Area Chair at AAAI 2012, NIPS 2012, ICML 2013+14 and RSS 2013+14.

See cool Robot projects and the videos from CS4758!. Described in Cornell Chronicle article. (See video. see photos)

Ashutosh Saxena is a Sloan Research Fellow. (New York Times.)

See all news.