AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

Refined bounds for algorithm configuration: The knife-edge of dual class approximability

Maria-Florina Balcan, Tuomas Sandholm, Ellen Vitercik

ICML 2020 · Learning theory and algorithms › Sample complexity · approach: PAC learnability theory · 3 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
Approximate dual performance functions under L∞ using a simpler function classAlgorithm configuration generalization bounds · General algorithm configurationgeneralization guaranteeImproved
Rely on finite-Lp approximation instead of L∞ approximationAlgorithm configuration generalization bounds · Worst casegeneralization guaranteeHurt
Use L∞ dual-function approximation with structural risk minimizationCPLEX variable-selection policy configuration · CATS arbitrary, score_L/score_Ssample complexity for generalization guarantee 0.1: 10^5 versus 7 \cdot 10^7; 700 timesImproved
Choose approximation complexity by structural risk minimization rather than fixing it to the maximum number of observed pieces (ablation)CPLEX variable-selection policy configuration · CATS integer programming experimentssample complexity: up to 4.5 times smallerImproved
Use L∞ approximation with structural risk minimization rather than the maximum-piece baseline (ablation)CPLEX variable-selection policy configuration · CATS arbitrary, score_L/score_Ssamples for generalization error 0.25: 6700 samples to 1500 samples; 4.6x improvementImproved
Use L∞ approximation with structural risk minimization rather than the maximum-piece baseline (ablation)CPLEX variable-selection policy configuration · CATS integer programming experimentssamples for generalization error 0.3: 4300 samples to 1400 samples; 3.07x improvementImproved

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