AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Learning Overcomplete, Low Coherence Dictionaries with Linear Inference

Jesse A. Livezey, Alejandro F. Bujan, Friedrich T. Sommer

JMLR 2019 · Probabilistic and scientific modeling › Molecules and proteins · approach: Sparse graphical and tensor models

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
replace L2 coherence cost with L4 costovercomplete ICA · 2x overcomplete dictionary, 32-dim, numerical minimizationpairwise angle distribution / coherenceImproved
use L2 cost from noisy pathological initializationovercomplete ICA coherence control · 2x overcomplete, 32-dimcoherenceHurt
remove quadratic Taylor term from Coulomb and Random Prior costs (flattened versions) (ablation)Coulomb and Random Prior costs · 2x overcomplete, 32-dim, 10 random initscoherence / pairwise angles near 90 degreesImproved
use Coulomb or Random Prior cost instead of L2 at overcompleteness above 2coherence control cost · 32-dim, varying overcompletenessmedian minimum pairwise angleImproved
use L2 cost or Score Matching instead of other ICA costsovercomplete ICA mixing matrix recovery · 32-dim, k=12 sparse, 2x overcompletenormalized recovery errorHurt
use sparse coding instead of ICA with linear inference at high overcompletenessmixing matrix recovery · 32-dim, k=12, large overcompletenessnormalized recovery errorImproved
use L2 cost on natural image patches2x overcomplete ICA · Van Hateren 8x8 whitened patchesduplicate bases / coherenceHurt

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