Reproduce “Efficient Tensor Kernel methods for sparse regression” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Feliks Hibraj, Marcello Pelillo, Saverio Salzo, Massimiliano Pontil
The authors reduce memory use in sparse tensor-kernel regression. Their layout stores only distinct values created by tensor symmetry, while a Nyström-like strategy fits each model on a random subset of training examples. At 250 points, a fourth-order tensor uses 1.24 GB rather than 29.1 GB under the assumed 8-byte doubles; these are storage estimates, not total memory.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.