Embedded Files
Today's methods for training artificial agents learn akin to agents that are locked alone in a room - a room full of books. Driven by large amounts of carefully selected and curated training data, machine learning has produced rapid progress in many tasks ranging from image understanding to natural language processing. While the resulting intelligent agent often demonstrates strong test set performance, agents still struggle when confronted with novel situations - this becomes particularly apparent in robotics, as it requires prediction and realtime adaptation to the curernt situation. We tackle inteactive tasks ranging from grasp pose synthesis to dynamic grasping of moving objects.
Amogh Tiwari, Karteek Alahari, and P. Bideau
Decoupling Object and grasp Representation Learning for Generalizable Grasp Pose Synthesis
International Workshop on AI for Robotics, 20.-21. Nov. 2025
Naver Labs, Grenoble, France
poster
Iñaki Román Martinez, Bilal Yağiz Gündeger, and Pia Bideau
Reinforcement Learning for Dynamic Object Grasping with Planner and Sensing Delays
International Workshop on AI for Robotics, 20.-21. Nov. 2025
Naver Labs, Grenoble, France
poster
Iñaki Román Martinez
Autonomous Robot Grasping in Motion using Reinforcement Learning
M.Sc. Mobile, Autonomous and Robotic Systems (MARS)
successfully defended: July 8th, 2025
supervisor: Pia Bideau
examiner: Ahmad Hably
M. Halawa, O. Hellwich, and P. Bideau
Action-Based Contrastive Learning for Trajectory Prediction
European Conference on Computer Vision (ECCV), Oct. 2022
DOI: 10.1007/978-3-031-19842-7_9
Page updated
Google Sites
Report abuse