English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Given the recent impressive accomplishments of language models (LMs) for codegeneration, we explore the use of LMs as adaptive mutation and crossoveroperators for an evolutionary neural architecture search (NAS) algorithm. WhileNAS still proves too difficult a task for LMs to succeed at solely throughprompting, we find that the combination of evolutionary prompt engineering withsoft prompt-tuning, a method we term EvoPrompting, consistently finds diverseand high performing models. We first demonstrate that EvoPrompting is effectiveon the computationally efficient MNIST-1D dataset, where EvoPrompting producesconvolutional architecture variants that outperform both those designed byhuman experts and naive few-shot prompting in terms of accuracy and model size.We then apply our method to searching for graph neural networks on the CLRSAlgorithmic Reasoning Benchmark, where EvoPrompting is able to design novelarchitectures that outperform current state-of-the-art models on 21 out of 30algorithmic reasoning tasks while maintaining similar model size. EvoPromptingis successful at designing accurate and efficient neural network architecturesacross a variety of machine learning tasks, while also being general enough foreasy adaptation to other tasks beyond neural network design.

EvoPrompting: Language Models for Code-Level Neural Architecture Search