MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chan, Jun Shern, Chowdhury, Neil, Jaffe, Oliver, Aung, James, Sherburn, Dane, Mays, Evan, Starace, Giulio, Liu, Kevin, Maksin, Leon, Patwardhan, Tejal, Weng, Lilian, Mądry, Aleksander
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929732112613376
author Chan, Jun Shern
Chowdhury, Neil
Jaffe, Oliver
Aung, James
Sherburn, Dane
Mays, Evan
Starace, Giulio
Liu, Kevin
Maksin, Leon
Patwardhan, Tejal
Weng, Lilian
Mądry, Aleksander
author_facet Chan, Jun Shern
Chowdhury, Neil
Jaffe, Oliver
Aung, James
Sherburn, Dane
Mays, Evan
Starace, Giulio
Liu, Kevin
Maksin, Leon
Patwardhan, Tejal
Weng, Lilian
Mądry, Aleksander
contents We introduce MLE-bench, a benchmark for measuring how well AI agents perform at machine learning engineering. To this end, we curate 75 ML engineering-related competitions from Kaggle, creating a diverse set of challenging tasks that test real-world ML engineering skills such as training models, preparing datasets, and running experiments. We establish human baselines for each competition using Kaggle's publicly available leaderboards. We use open-source agent scaffolds to evaluate several frontier language models on our benchmark, finding that the best-performing setup--OpenAI's o1-preview with AIDE scaffolding--achieves at least the level of a Kaggle bronze medal in 16.9% of competitions. In addition to our main results, we investigate various forms of resource scaling for AI agents and the impact of contamination from pre-training. We open-source our benchmark code (github.com/openai/mle-bench/) to facilitate future research in understanding the ML engineering capabilities of AI agents.
format Preprint
id arxiv_https___arxiv_org_abs_2410_07095
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering
Chan, Jun Shern
Chowdhury, Neil
Jaffe, Oliver
Aung, James
Sherburn, Dane
Mays, Evan
Starace, Giulio
Liu, Kevin
Maksin, Leon
Patwardhan, Tejal
Weng, Lilian
Mądry, Aleksander
Computation and Language
We introduce MLE-bench, a benchmark for measuring how well AI agents perform at machine learning engineering. To this end, we curate 75 ML engineering-related competitions from Kaggle, creating a diverse set of challenging tasks that test real-world ML engineering skills such as training models, preparing datasets, and running experiments. We establish human baselines for each competition using Kaggle's publicly available leaderboards. We use open-source agent scaffolds to evaluate several frontier language models on our benchmark, finding that the best-performing setup--OpenAI's o1-preview with AIDE scaffolding--achieves at least the level of a Kaggle bronze medal in 16.9% of competitions. In addition to our main results, we investigate various forms of resource scaling for AI agents and the impact of contamination from pre-training. We open-source our benchmark code (github.com/openai/mle-bench/) to facilitate future research in understanding the ML engineering capabilities of AI agents.
title MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering
topic Computation and Language
url https://arxiv.org/abs/2410.07095