Deep SE(3)-Equivariant Geometric Reasoning for Precise Placement Tasks

no code implementations • 20 Apr 2024 • Ben Eisner, Yi Yang, Todor Davchev, Mel Vecerik, Jonathan Scholz, David Held

In this work, we propose a method for precise relative pose prediction which is provably SE(3)-equivariant, can be learned from only a few demonstrations, and can generalize across variations in a class of objects.

Pose Prediction Position +1

RoboTAP: Tracking Arbitrary Points for Few-Shot Visual Imitation

no code implementations • 30 Aug 2023 • Mel Vecerik, Carl Doersch, Yi Yang, Todor Davchev, Yusuf Aytar, Guangyao Zhou, Raia Hadsell, Lourdes Agapito, Jon Scholz

For robots to be useful outside labs and specialized factories we need a way to teach them new useful behaviors quickly.

RoboCat: A Self-Improving Generalist Agent for Robotic Manipulation

no code implementations • 20 Jun 2023 • Konstantinos Bousmalis, Giulia Vezzani, Dushyant Rao, Coline Devin, Alex X. Lee, Maria Bauza, Todor Davchev, Yuxiang Zhou, Agrim Gupta, Akhil Raju, Antoine Laurens, Claudio Fantacci, Valentin Dalibard, Martina Zambelli, Murilo Martins, Rugile Pevceviciute, Michiel Blokzijl, Misha Denil, Nathan Batchelor, Thomas Lampe, Emilio Parisotto, Konrad Żołna, Scott Reed, Sergio Gómez Colmenarejo, Jon Scholz, Abbas Abdolmaleki, Oliver Groth, Jean-Baptiste Regli, Oleg Sushkov, Tom Rothörl, José Enrique Chen, Yusuf Aytar, Dave Barker, Joy Ortiz, Martin Riedmiller, Jost Tobias Springenberg, Raia Hadsell, Francesco Nori, Nicolas Heess

With RoboCat, we demonstrate the ability to generalise to new tasks and robots, both zero-shot as well as through adaptation using only 100-1000 examples for the target task.

Wish you were here: Hindsight Goal Selection for long-horizon dexterous manipulation

no code implementations • ICLR 2022 • Todor Davchev, Oleg Sushkov, Jean-Baptiste Regli, Stefan Schaal, Yusuf Aytar, Markus Wulfmeier, Jon Scholz

In this work, we extend hindsight relabelling mechanisms to guide exploration along task-specific distributions implied by a small set of successful demonstrations.

Continuous Control Reinforcement Learning (RL)

Learning Time-Invariant Reward Functions through Model-Based Inverse Reinforcement Learning

no code implementations • 7 Jul 2021 • Todor Davchev, Sarah Bechtle, Subramanian Ramamoorthy, Franziska Meier

Inverse reinforcement learning is a paradigm motivated by the goal of learning general reward functions from demonstrated behaviours.

reinforcement-learning Reinforcement Learning (RL)

Model-Based Inverse Reinforcement Learning from Visual Demonstrations

no code implementations • 18 Oct 2020 • Neha Das, Sarah Bechtle, Todor Davchev, Dinesh Jayaraman, Akshara Rai, Franziska Meier

Scaling model-based inverse reinforcement learning (IRL) to real robotic manipulation tasks with unknown dynamics remains an open problem.

Model Predictive Control reinforcement-learning +1

Residual Learning from Demonstration: Adapting DMPs for Contact-rich Manipulation

no code implementations • 18 Aug 2020 • Todor Davchev, Kevin Sebastian Luck, Michael Burke, Franziska Meier, Stefan Schaal, Subramanian Ramamoorthy

Dynamic Movement Primitives (DMP) are a popular way of extracting such policies through behaviour cloning (BC) but can struggle in the context of insertion.

Behavioural cloning Friction +1

Learning Structured Representations of Spatial and Interactive Dynamics for Trajectory Prediction in Crowded Scenes

1 code implementation • 29 Nov 2019 • Todor Davchev, Michael Burke, Subramanian Ramamoorthy

Context plays a significant role in the generation of motion for dynamic agents in interactive environments.

motion prediction Robot Manipulation +1

An Empirical Evaluation of Adversarial Robustness under Transfer Learning

no code implementations • 7 May 2019 • Todor Davchev, Timos Korres, Stathi Fotiadis, Nick Antonopoulos, Subramanian Ramamoorthy

Specifically, we study the effects of using robust optimisation in the source and target networks.

Adversarial Robustness Transfer Learning