Ashutosh Saxena

Assistant Professor,
Department of Computer Science,
Cornell University.
Education: PhD, Stanford University.

4159 Upson 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.

Personal Robotics
Make3D
Aerial Robotics


How can robots operate autonomously in unstructured environments? I design learning algorithms to help robots perceive the environments so that they can perform tasks such as open doors, pick up novel objects, unload items from a dishwasher, fly around without crashing into obstacles, and more.

I am also interested in developing new machine learning algorithms, especially ones that can handle high-dimensionality of the perceptual space (e.g., CRFs), while also considering the actions in time (e.g., reinforcement learning).


Getting involved:

Undergraduates at Cornell: Please first send me an email with your resume and transcript; you must know C, C++ or C# and have a GPA of 3.7 or above. Sophomores and juniors CS/ECE are especially encouraged to apply.

PhD students (CS, ECE and MAE) at Cornell: Please email me for an appointment!

Students not at Cornell: Please go through our admissions first.

Postdoc position available: vision for robotics, machine learning, aerial robotics.

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

Yun Jiang's work on picking up and arranging objects was featured in several news articles.

Jiang/Lim's works on learning arrangement of objects in 3D scene were accepted to ICML and ISER conferences.

Yun Jiang's work in placing new objects was accepted in IJRR'12.

Jiang/Amend's work featured in Cornell Chronicle.

Dhruv Batra's work on learning Laplacian CRFs was accepted in CVPR'12.

Prof. Saxena is co-organizing manipulation workshop at IROS 2012.

Anand/Koppula release new code and Cornell RGBD dataset for indoor scene labeling.

Daniel Ly's paper on fitting arbitrary kinematic skeletons to RGB-D data is accepted in GECCO'12.

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 and NIPS 2012.

Three papers (Sung/Ponce's on activity detection, Jiang/Amend's on grasping with deformable gripper, and Jiang/Lim/Zheng's on placing objects) and Jiang/Basu's video accepted to ICRA 2012.

Anand/Koppula and Congcong Li presented their respective papers at NIPS 2011.

Anand/Koppula's work featured in New Scientist. Jae Sung's work featured in R&D Magazine.

Li/Kowdle's paper on FeCCM was accepted in IEEE TPAMI.

Jiang/Zheng/Lim, Labutov/Yosinski, Yang/Low/Cong, Ly, and Anand/Koppula to present their respective works at the RSS workshops in Los Angeles.

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.)

Cornell's Personal Robotics group in the Communications of the ACM and others.

Congcong Li, TP Wong and Norris Xu build a "shoe finder robot" in a day using FeCCMs, featured in ICRA 2011 videos. (More) Also mentioned in New Scientist.

See all news.