AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

TabNAS: Rejection Sampling for Neural Architecture Search on Tabular Datasets

Chengrun Yang, Gabriel Bender, Hanxiao Liu, Pieter-Jan Kindermans, Madeleine Udell, Yifeng Lu, Quoc V Le, Da Huang

NeurIPS 2022 · Reinforcement learning › Policy gradient and actor-critic · approach: Alternative policy update rules · 1 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 Abs Reward with rejection-based reward (ablation)RL weight-sharing NAS · Criteo 3-layervalidation loss: 0.4461 -> 0.4455Improved
replace Abs Reward with rejection-based reward (ablation)RL weight-sharing NAS · Criteo 4-layervalidation loss: 0.4452 -> 0.4449Improved
replace Abs Reward with rejection-based reward (ablation)RL weight-sharing NAS · Criteo 5-layervalidation loss: 0.4448 -> 0.4445Improved
replace Abs Reward with rejection-based reward (ablation)RL weight-sharing NAS · Aloibalanced error: 0.0662 -> 0.0669No clear effect
replace Abs Reward with rejection-based reward (ablation)RL weight-sharing NAS · Criteo 4-layerhyperparameter settings finding reference: 3 out of 338Improved
replace MnasNet reward with rejection-based reward (ablation)RL weight-sharing NAS · tabular search spacesfraction of trials within 5% of target size: fewer than 1%Improved
use larger number of Monte-Carlo samples in rejection estimate (ablation)TabNAS · Criteo 3-layerfinding reference architecture: no fewer than 2,048Improved
use bottleneck architecturefeedforward network · Aloibalanced error: 0.0662 -> 0.0669No clear effect
vary beta in Abs Reward (small versus large magnitude) (ablation)Abs Reward NAS · Volkert 4-layer / toy examplefeasibility and qualityMixed
replace Abs Reward with rejection-based reward (ablation)RL weight-sharing NAS · NATS-Bench size search spaceaccuracyImproved
train SuperNet without updating the RL controller (ablation)SuperNet one-shot weights · Criteo 3-layercorrelation of one-shot and stand-alone loss: 0.96No 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