AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Target Curricula via Selection of Minimum Feature Sets: a Case Study in Boolean Networks

Shannon Fenn, Pablo Moscato

JMLR 2017 · Learning theory and algorithms › Prediction and decision losses · approach: Loss landscape analysis · 4 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
replace L1 loss with strict greatest-hierarchy loss L_ghFBN with NAND gates · 7-bit cascaded parity, 6th targettest accuracy: 58% -> 84%Improved
use L_gh instead of L1FBN with NAND gates · 7-bit cascaded paritytest accuracy: 16% reduction in training examples to exceed 90% mean test accuracyImproved
order targets by easy-to-hard strictness (L_w -> L_lh -> L_gh) (ablation)FBN target curriculum · five test-bed problemstest accuracyImproved
use automatically detected target order instead of known orderL_gh · test-bed problems except CMUXtest accuracyNo clear effect
use automatically detected target orderL_gh · CMUXtest accuracyMixed
reverse the target order (tau = -1) (ablation)L_gh · 5-bit additiontest accuracyHurt
use exact easy-to-hard order (tau = 1) over random orders (ablation)L_gh · 5-bit additiontest accuracyImproved
use random target order (ablation)L_gh · 5-bit additiontest accuracyNo clear effect
add hierarchical losses with detected orderL_w/L_lh/L_gh · mammalian cell-cycle modeltest accuracyMixed
use L_lh instead of L1FBN · E. coli target G8mean test accuracy difference vs L1: -0.9Hurt
use L_gh instead of L1 on later target in hierarchyFBN · E. coli target G6mean test accuracy difference vs L1: 1.4Improved
use L_gh instead of L1FBN · Yeast SK -> {Ste9, Rum1} hierarchymean test accuracy difference vs L1: 0.9, 0.5, 0.3No clear effect

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