Irmak Guzey

PhD student in Computer Science, New York University (she/her)

profile_photo.jpg

irmakguzey@nyu.edu

New York, NY, United States

I am currently a first year PhD student in Computatinal Intelligence, Vision, and Robotics Lab (CILVR) at NYU and am lucky to be advised by Prof. Lerrel Pinto.

I had my master’s degree from New York University, Computer Science at Courant Institute of Mathematical Sciences. I was awarded with a Fulbright scholarship in 2020.

My research mostly focus on Policy Learning and Representation Learning. I enjoy working on tackling different tasks in Manipulation and Navigation in Robotics.

On top of my research, with my interest in Robotics and Machine Learning, I worked as a full-time robotics engineer in DOF Robotics for a year, had an internship in X - the Moonshot Factory and 2 different internships in Google, one of them being purely on machine learning.

News

Sep 1, 2024 Started my PhD program in Computer Science at NYU, Courant Institute of Mathematical Sciences! I will be working on Robot Learning for Dexterous Manipulation, advised by Lerrel Pinto.
Sep 25, 2023 We just published our latest project TAVI (See to Touch: Learning Tactile Dexterity through Visual Incentives)! It got accepted to ICRA 2024. TAVI learns an online visuo-tactile policy using image only guidance. Learned policy can adapt to new environments under only 1 hour of training! For more check out website and code.
Mar 22, 2023 T-Dex (Dexterity from Touch: Self-Supervised Pre-Training of Tactile Representations with Robotic Play) got accepted to CoRL 2023! This is the first conference publication of mine as the main author. T-Dex is a dexterous manipulation framework that uses tactile observations. Tactile encoders are trained using SSL objectives on a tactile play data.
Oct 12, 2022 Holo-Dex: Teaching Dexterity with Immersive Mixed Reality got accepted to ICRA 2023! This is the first publication of my master’s and I had a blast working on this project with my colleagues. Holo-Dex is also out on Arxiv now.
Sep 1, 2021 Started my master’s program in Computer Science in NYU, Courant Institute of Mathematical Sciences!
Sep 1, 2021 Ended my one year of full-time working experience as a robotics engineer in DOF robotics to start my master’s program. Worked on implementing an autonomous guided vehicle (AGV) using ROS.
Nov 30, 2020 Ended my internship as a Machine Learning Engineer in Google, Paris. My main work was focused on improving Youtube’s annotation system by feature engineering.
Aug 15, 2020 Was awarded with a Fulbright scholarship on Computer Science. Was one of the 3 people who was granted a Fulbright scholarship in Computer Science in Turkey.
Jun 2, 2020 My first publication Human Motion Prediction with Graph Neural Networks was accepted to Long-Term Human Motion Prediction workshop of ICRA 2020 !.
Sep 10, 2019 Ended my internship as a Software Engineer in X, the Moonshot Factory.
Aug 20, 2018 Ended my internship as a Software Engineer in Google Zurich! Worked on integrating UEFI Extraction’s results into the pipelines of malware detection systems.

Selected Publications

  1. HuDOR: Bridging the Human to Robot Dexterity Gap through Object-Oriented Rewards
    Irmak Guzey, Yinlong Dai, Georgy Savva, Raunaq Bhirange, and Lerrel Pinto
    (Under Review) 2024
  2. See to Touch: Learning Tactile Dexterity through Visual Incentives
    Irmak Guzey, Yinlong Dai, Ben Evans, Soumith Chintala, and Lerrel Pinto
    ICRA 2024 2023
  3. Dexterity from Touch: Self-Supervised Pre-Training of Tactile Representations with Robotic Play
    Irmak Guzey, Ben Evans, Soumith Chintala, and Lerrel Pinto
    CoRL 2023 2023
  4. Holo-Dex: Teaching Dexterity with Immersive Mixed Reality
    Sridhar Pandian Arunachalam, Irmak Guzey, Soumith Chintala, and Lerrel Pinto
    ICRA 2023 2022
  5. Human Motion Prediction With Graph Neural Networks
    Irmak Guzey, Ahmet E. Tekden, Evren Samur, and Emre Ugur
    2020