JEPA Pointed Out as Similar to PMAX Concepts
SchmidhuberAI · x · 2026-07-18
This in-depth post discusses the relationship between LeCun's JEPA and Schmidhuber's team 1992 PMAX, with the core argument being: the key ideas behind JEPA are not new.
The author suggests that PMAX is essentially a method for two non-generative neural networks to learn latent representations through prediction/matching. They further point out that JEPA is highly similar to this concept, providing extensive references to support their claim.
More from Research
- Structural ensembles beat single predictions in TCR:pMHC generalization study — quaidmorris · 2026-07-22
- Structural ensembles, not single predictions, drive robust TCR:pMHC generalization — quaidmorris · 2026-07-22
- enFoldX turns AlphaFold3 ensemble noise into a TCR–peptide–MHC predictor — quaidmorris · 2026-07-22
- RSS launches under OMSF to push structural biology data modeling at scale — MoAlQuraishi · 2026-07-22
- enFoldX tops 8 neoantigen scans and an unseen-peptide benchmark — quaidmorris · 2026-07-22
- enFoldX reaches AUC 0.82 on human VDJdb and transfers to mouse at 0.76 — quaidmorris · 2026-07-22