# PredNAS: A Universal and Sample Efficient Neural Architecture Search Framework

Source: [PredNAS: A Universal and Sample Efficient Neural Architecture Search Framework](<https://www.sota2.com/research/paper/68fa6dc51e209131cea3c091>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:56:55.986Z

Authors: Liuchun Yuan, Zehao Huang, Naiyan Wang

Published: 2022-10-26T04:15:58.000Z

## Abstract

In this paper, we present a general and effective framework for Neural Architecture Search (NAS), named PredNAS. The motivation is that given a differentiable performance estimation function, we can directly optimize the architecture towards higher performance by simple gradient ascent. Specifically, we adopt a neural predictor as the performance predictor. Surprisingly, PredNAS can achieve state-of-the-art performances on NAS benchmarks with only a few training samples (less than 100). To validate the universality of our method, we also apply our method on large-scale tasks and compare our method with RegNet on ImageNet and YOLOX on MSCOCO. The results demonstrate that our PredNAS can explore novel architectures with competitive performances under specific computational complexity constraints.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2210.14460>)

[PDF](<https://arxiv.org/pdf/2210.14460.pdf>)