Yujia Zheng's banner

Yujia Zheng

@YujiaZheng9 • 1,738 subscribers

Assistant Professor @UofIllinois. PhD @CarnegieMellon. Interested in causal learning and machine learning.

Shorts

Is it possible to provably recover individual latent variables of the true world, even without reconstruction (e.g., JEPA)? Yes, with DSReg, a simple regularization that can be applied post hoc to your pretrained model!

Is it possible to provably recover individual latent variables of the true world, even without reconstruction (e.g., JEPA)? Yes, with DSReg, a simple regularization that can be applied post hoc to your pretrained model!

40,436 Aufrufe