AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

Reverse-mode differentiation in arbitrary tensor network format: with application to supervised learning

Alex A. Gorodetsky, Cosmin Safta, John D. Jakeman

JMLR 2022 · Learning theory and algorithms › Matrix and tensor methods · approach: Tensor network decompositions · 1 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 gradient-based nonlinear least squares instead of ALS to fit the tensor networktensor network recovery (5x5 matrix, rank 2) · two-dimensional function recoveryprobability of recovery: recovery using NLS occurs starting at 15 samples while ALS is delayed until 20 samplesImproved
use NLS instead of ALS on an over-parameterized 3x3 matrix, rank 3tensor network recovery · two-dimensional function recoveryprobability of recovery: both approaches recover the function once nine data points are obtainedNo clear effect
use NLS instead of ALS on an over-parameterized 5x5 matrix, rank 4tensor network recovery · two-dimensional function recoveryprobability of recovery: NLS immediately fits well once the 25 sample threshold is passedImproved
use NLS instead of ALS on balanced tree network (pi=2, r=2)tree-tensor network · four-dimensional function recoveryprobability of recovery: both ALS and NLS converge almost immediately after observing 16 data pointsNo clear effect
use NLS instead of ALS on balanced tree network (pi=3, r=3)tree-tensor network · four-dimensional function recoverysamples needed for recovery: NLS does seem to converge 10 samples earlier than the ALSImproved
use NLS instead of ALS on balanced tree network (pi=4, r=2, 52 parameters)tree-tensor network · four-dimensional function recoveryprobability of recovery: earlier transition into convergence by NLS, around approximately 50 samplesImproved
use NLS instead of ALS on non-traditional 5-D network (410 unknowns)non-traditional tensor network · five-dimensional function recoverysamples needed for recovery: begins to converge almost one hundred samples earlier than the ALS solverImproved
use NLS instead of ALS on 16-D unbalanced tree network (745 parameters)non-traditional tensor network · sixteen-dimensional function recoverysamples needed for recovery: ALS needs several more times the number of samples to converge; ALS only converges to a level of 10-5Improved
use hierarchical Tucker with gradient-based fitting instead of functional tensor trainregression on real datasets · Housing, Speech, Music, Airfoil, Forestfiresvalidation MSE: Music 0.72094 -> 0.6536; Airfoil 0.46920 -> 0.40708; Forestfires 0.39465 -> 0.352Improved
use hierarchical Tucker with gradient-based fitting instead of functional tensor trainregression on real datasets · Skillcraft, Telemonit, Propulsionvalidation MSE: Telemonit 0.04040 -> 0.079516; Propulsion 0.00009 -> 0.0006952; Skillcraft 0.54434 -> 0.70944Hurt
use hierarchical Tucker with gradient-based fitting instead of functional tensor trainregression on real datasets · CCPP (d=59, n=2000)validation MSE: 0.06631 -> 0.06843No clear effect
use hierarchical Tucker with gradient-based fitting instead of functional tensor trainregression on real datasets · Galaxy (d=20, n=2000)validation MSE: 0.00056 -> 0.000017Improved

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