AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Learning the Efficient Frontier

Philippe Chatigny, Ivan Sergienko, Ryan Ferguson, Jordan Weir, Maxime Bergeron

NeurIPS 2023 · Learning theory and algorithms › Combinatorial and search optimization · approach: Constrained neural network optimization · 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
Replace numerical EF optimization with NeuralEF inferenceEF optimization · EF throughput, A100evaluations/second: 111479.39 -> 343760.77Improved
Use FP16 instead of FP32 (ablation)NeuralEF · EF throughput, A100evaluations/second: 128950.46 -> 343760.77Improved
Use FP16 instead of FP32 (ablation)NeuralEF · EF accuracyportfolio weight error: The paper reports FP32 and FP16 results in separate tables, without a direct paired delta.No clear effect
Add dynamic greedy allocation rebalancing (DGAR) (ablation)NeuralEF · EF test distributionportfolio weight MAEImproved
Add dynamic greedy allocation rebalancing (DGAR) (ablation)NeuralEF · EF test distributionconstraint adherence optimality gapImproved
Clean ambiguous optimization inputs before token projection (ablation)NeuralEF · EF inputs with equivalent but ambiguously specified constraintsforecast accuracyImproved
Remove input preprocessing (ablation)NeuralEF · EF throughput, A100evaluations/second: 343760.77 -> 401641.81Improved
Remove input preprocessing (ablation)NeuralEF · EF forecast accuracy on ambiguous inputsforecast accuracyHurt
Use clean-only input preprocessing (ablation)NeuralEF · EF throughput, A100evaluations/second: 343760.77 -> 366450.96Improved
Train and evaluate within the training input domainNeuralEF · EF out-of-domain inputstail forecast accuracyImproved

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