Predictive Architecture Papers

CVPR 2023

I-JEPA: Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture

Mahmoud Assran, Quentin Duval, +6

How I-JEPA learns visual representations by predicting abstract feature representations of masked image regions — no pixel reconstruction, no augmentation — achieving 81.7% linear probe accuracy with ViT-H.

Predictive Architecture
Self Supervised Learning
Joint Embedding
+1 more tags

Concepts

Paper Code

Self Attention ViT
Collapse Risk
MSE MAE

Predictive Architecture

Papers Related to Predictive Architecture

I-JEPA: Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture

Concepts