AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Learning dynamics of deep linear networks with multiple pathways

Jianghong Shi, Eric Todd SheaBrown, Michael A Buice

NeurIPS 2022 · Architectures and efficiency › RNNs and normalization · approach: Brain-inspired network comparison · 2 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
increase pathway depth from D = 2 to D = 7 (ablation)two-pathway deep linear network · 8-dim identity input, 15-dim output simulationconcentration of each singular value in one pathway: D = 2 -> D = 7Improved
use depth 3 and 4 instead of depth 2 (ablation)two-pathway deep linear network · 8-dim identity input, 15-dim output simulationsymmetry between pathways: D = 2 -> D = 3, 4Improved
use ReLU nonlinearity after each layer instead of linear (ablation)two-pathway network · depths 2, 3, 4, 7 simulationfeature concentration in one pathwayNo clear effect
use Tanh nonlinearity after each layer instead of linear (ablation)two-pathway network · depths 2, 3, 4, 7 simulationfeature concentration in one pathwayNo clear effect
larger initial weight fluctuations (sigma = 0.1 versus 0.01) (ablation)two-pathway phase portrait, single alpha · S_alpha = 10 phase portraittrajectory drift toward one dominant pathway: sigma = 0.01 -> sigma = 0.1Improved
increase hidden width to the infinite limitmulti-pathway linear network · analytic resultsharing of features across pathwaysImproved
different random seed for initialization (ablation)two-pathway network D = 7 · simulation, Na = Nb = 1000, sigma = 0.01which pathway holds the largest singular valuesMixed

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