Vincent-Pierre Berges

Research Engineer @metaAI

[AI Habitat @ai_habitat]

Announcing: Habitat Rearrangement Challenge at NeurIPS 2022!
Goal: Control a home assistant robot and rearrange 1 object from start to goal position.

[Vincent-Pierre Berges @vinceberges]

@__ReJ__ @_eriq_ @_bitter_ @wumpf @sielojramu @FlorentGuinier Keep being awesome man. I wish you the best on your next endeavor!

[Vincent-Pierre Berges @vinceberges]

@formalsystem @awjuliani Hey!
This page might help https://t.co/EhZjZqpiL9
You can enumerate the keys in the registry with something like :
list(default_registry.keys())

[Vincent-Pierre Berges @vinceberges]

The latest release of #mlagents introduces the #unity3d Environment Registry! This allows you to use our example environments for #machinelearning anytime and anywhere.
Here is a notebook running a simple Q-Learning on the GridWorld environment! https://t.co/MxyWvUeLmU

[Danny Lange @danny_lange]

Today, after more than two and a half years of development and 15+ release updates, we are excited to announce that the ML-Agents Unity package has reached v1.0: https://t.co/H2D0aEgRiI @unity3d #AI #Simulation

[Unity for Games @unitygames]

After 2.5 amazing years working with studios, researchers, and students all over the world, we are excited to announce that #MLAgents Unity Package v1.0 is now available as a preview package! 🎉

Learn more about this release: https://t.co/U9dZLISU9o

[Digital Opus @DigitalOpus]

Added a reward for matching a pose and got a much better result. Although I accidentally set the target height too high. To get the reward it is standing on its tippy toes. #mlagents

[Andrew Cohen @andrew_e_cohen]

Check out my new post on the @unity3d blog! Learn all about our new self-play feature that’s part of the 0.14 release of #MLAgents https://t.co/nrWHHKTqho #unity #unity3d

[AI and Games School 2026 @GameAISchool]

Meet guest lecturer Vincent-Pierre Berges @vinceberges Senior MachineLearning Engineer at @unity3d! He was one of the founders of Unity MachineLearning Agents and joins #GameAISchool2020 with a workshop on Unity and ML for Research & NPC AI!
Register now! https://t.co/wGMZOIqhMn

[Vincent-Pierre Berges @vinceberges]

Me : Okay, now don't go around the wall, that's cheating.
My reinforcement learning agent : Okay then.

[Vincent-Pierre Berges @vinceberges]

Me : Use that block to scale the wall.
My reinforcement learning agent : Why?

[Unity for Games @unitygames]

ML-Agents v0.14 is here! What’s included?

✅ A new self-play feature for adversarial scenario
✅ Unity Package for #MLAgents
✅ Several refactoring improvements
& much more!

See what's new with the release on GitHub. https://t.co/R4Utgyb3gc

[Andy Wojcicki @pretendsmarts]

@unity3d and ML-agents remain the heart of the Arena Building Toolkit, so big thanks to @awjuliani @vinceberges @shihzy and all the ML folks at Unity, as well as @MaCroPhilosophy and @BenBeyret for paving the way with Animal-AI Olympics and providing great feedback

[Vincent-Pierre Berges @vinceberges]

Playing around with reward functions gone wrong. #MachineLearning