Home▸AI & Machine Learning▸2D LoRA Low-Rank Adapter: Eckart-Young Rank Limit, Gradient Descent on B and A, and Old-Task Drift

🧩 2D LoRA Low-Rank Adapter: Eckart-Young Rank Limit, Gradient Descent on B and A, and Old-Task Drift

Heat maps of the target weight update, the learned low-rank update and the residual, with the singular spectrum split into kept and lost directions: gradient descent on B and A reaches the Eckart-Young floor, with finite-difference gradient and null-space checks.

AI & Machine Learning2DAdvanced60 FPS📱 Mobile-adapted⇄ 3D version
2D LoRA Low-Rank Adapter: Eckart-Young Rank Limit, Gradient Descent on B and A, and Old-Task Drift ↗ Open standalone
⚙ Under the hood

Heat maps of the target weight update, the learned low-rank update and the residual, with the singular spectrum split into kept and lost directions: gradient descent on B and A reaches the Eckart-Young floor, with finite-difference gradient and null-space checks.

LoRALow-Rank AdaptationFine-TuningEckart-YoungSingular ValuesGradient DescentCatastrophic ForgettingParameter-Efficient

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)