AI Paper Atlas

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

Delegated Classification

Eden Saig, Inbal Talgam-Cohen, Nir Rosenfeld

NeurIPS 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Active learning and subset selection · 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
increase validation set size mmin-budget contracts · MNIST, target accuracy 0.85required budget B*Improved
switch from MLP to GBDT as delegated learner at accuracy ~0.74delegated learning · MNISTrequired budget: DeltaB* = -10^2Improved
switch from MLP to GBDT as delegated learner at accuracy ~0.94delegated learning · MNISTrequired budget: DeltaB* = 10^5Hurt
restrict optimization to simple contracts instead of general min-budgetcontract design · LCDB multiple learnersrequired budget: less than 1% higherNo clear effect
try SBA before generic LP solvermin-budget contract solver · LCDBsolve time and success: more than 85% success, ~10^3 times fasterImproved
low bias (larger n0, r=1) instead of low variance (more samples per n) for curve fitlearning curve extrapolation · MNIST partial informationaccuracy via delegationImproved
increase pilot set size klearning curve estimation · MNIST partial informationaccuracy obtained via delegationImproved
use r=1 rather than r>1 samples per nsample-size multiplier · MNIST partial informationmu(k): ~10 -> ~30Improved
over-estimate learning curve instead of under-estimatecontract from extrapolated curve · MNIST partial informationaccuracy loss from delegationHurt
under-estimate learning curvecontract from extrapolated curve · MNIST partial informationaccuracy loss from delegationNo clear effect
relax C-MLRP assumption for threshold contractsthreshold contracts · LCDBoptimality of threshold formNo 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