no code implementations • 12 Feb 2024 • Pierre Marza, Laetitia Matignon, Olivier Simonin, Christian Wolf
We evaluate the method on a wide variety of tasks from the CortexBench benchmark and show that, compared to existing work, it can be addressed with a single policy.
1 code implementation • 21 Apr 2023 • Pierre Marza, Laetitia Matignon, Olivier Simonin, Dhruv Batra, Christian Wolf, Devendra Singh Chaplot
Empirical results show that NeRFs can be trained on actively collected data using just a single episode of experience in an unseen environment, and can be used for several downstream robotic tasks, and that modular trained exploration models outperform other classical and end-to-end baselines.
1 code implementation • ICCV 2023 • Pierre Marza, Laetitia Matignon, Olivier Simonin, Christian Wolf
Understanding and mapping a new environment are core abilities of any autonomously navigating agent.
no code implementations • 22 Dec 2021 • Edward Beeching, Maxim Peter, Philippe Marcotte, Jilles Debangoye, Olivier Simonin, Joshua Romoff, Christian Wolf
We address planning and navigation in challenging 3D video games featuring maps with disconnected regions reachable by agents using special actions.
1 code implementation • 7 Dec 2021 • Edward Beeching, Jilles Debangoye, Olivier Simonin, Christian Wolf
We present Godot Reinforcement Learning (RL) Agents, an open-source interface for developing environments and agents in the Godot Game Engine.
2 code implementations • 13 Jul 2021 • Pierre Marza, Laetitia Matignon, Olivier Simonin, Christian Wolf
In the context of visual navigation, the capacity to map a novel environment is necessary for an agent to exploit its observation history in the considered place and efficiently reach known goals.
1 code implementation • ECCV 2020 • Edward Beeching, Jilles Dibangoye, Olivier Simonin, Christian Wolf
We train an agent to navigate in 3D environments using a hierarchical strategy including a high-level graph based planner and a local policy.
no code implementations • 24 Jan 2020 • Edward Beeching, Christian Wolf, Jilles Dibangoye, Olivier Simonin
The EgoMap architecture incorporates several inductive biases including a differentiable inverse projection of CNN feature vectors onto a top-down spatially structured map.
1 code implementation • 3 Apr 2019 • Edward Beeching, Christian Wolf, Jilles Dibangoye, Olivier Simonin
In this paper we argue that research on training agents capable of complex reasoning can be simplified by decoupling from the requirement of high fidelity photographic observations.