- 🔭I am a PhD student at Volkswagen and TUM, working on world models focusing on pretraining and transfer.
- 🤔I am generally an AI enthusiast and interested in all related topics. I believe AGI can be achieved in my lifetime, and I am glad to be part of the journey.
- ⚡In my non-research life, I am a board game lover. If you want a match with me, you can find me with IcarusWizard on BGA.
- 📫If you want to contact me, please reach out to [email protected].
🤔
An intelligent machine who wants to understand himself. Doing PhD in world models.
-
Machine Learning Research Lab, Volkswagen AG
- Munich, Germany
- https://icaruswizard.github.io/
- @xyzhang42
- in/xingyuan-zhang
- https://scholar.google.com/citations?user=BUqNLqcAAAAJ
Highlights
- Pro
Pinned Loading
-
PixelwiseRegression
PixelwiseRegression PublicPyTorch release for paper "Pixel-wise Regression: 3D Hand Pose Estimation via Spatial-form Representation and Differentiable Decoder"
-
argmax-ai/aime
argmax-ai/aime PublicOfficial repository for our paper on "Action Inference by Maximising Evidence: Zero-Shot Imitation from Observation with World Models"
-
polixir/OfflineRL
polixir/OfflineRL PublicA collection of offline reinforcement learning algorithms.
-
polixir/NeoRL
polixir/NeoRL PublicPython interface for accessing the near real-world offline reinforcement learning (NeoRL) benchmark datasets
-
polixir/d3pe
polixir/d3pe PublicD3PE (Deep Data-Driven Policy Evaluation) aims to evaluation a large set of candidate policies from a fixed dataset to select best ones.
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.