AI Paper Atlas

Research map / Optimization / Optimizers

Near-Universal Multiplicative Updates for Nonnegative Einsum Factorization

John Hood, Aaron Schein

ICML 2026 · Optimization › Optimizers · approach: Surrogate losses and robust solvers · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use NNEinFact multiplicative updates instead of Adam automatic differentiationcustom nonnegative tensor models · Uber, ICEWS, WITS heldoutheldout loss and time to convergence: less than half the test loss, up to 90 times fasterImproved
use custom tailored decompositions instead of CP/Tucker/tensor-trainNNEinFact · Uber, ICEWS, WITS heldoutheldout (alpha,beta)-divergence: over 37% better predictionImproved
custom model vs CP at ICEWS alpha=0.8NNEinFact custom · ICEWS, alpha=0.8heldout loss: CP offers a 1% reductionHurt
multiplicative updates instead of HALS for CP and TuckerCP, Tucker · Uber, ICEWS, WITSheldout lossImproved
low-rank Tucker instead of full Tucker (ablation)Tucker · ICEWS, WITSheldout lossMixed
NNEinFact vs Adam on two of six (alpha,beta) settingsNNEinFact · Uberruntime and heldout loss: better on four of six settingsMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy