DS-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning

Siyuan Guo, Cheng Deng, Ying Wen, Hechang Chen, Yi Chang, Jun Wang
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:16813-16848, 2024.

Abstract

In this work, we investigate the potential of large language models (LLMs) based agents to automate data science tasks, with the goal of comprehending task requirements, then building and training the best-fit machine learning models. Despite their widespread success, existing LLM agents are hindered by generating unreasonable experiment plans within this scenario. To this end, we present DS-Agent, a novel automatic framework that harnesses LLM agent and case-based reasoning (CBR). In the development stage, DS-Agent follows the CBR framework to structure an automatic iteration pipeline, which can flexibly capitalize on the expert knowledge from Kaggle, and facilitate consistent performance improvement through the feedback mechanism. Moreover, DS-Agent implements a low-resource deployment stage with a simplified CBR paradigm to adapt past successful solutions from the development stage for direct code generation, significantly reducing the demand on foundational capabilities of LLMs. Empirically, DS-Agent with GPT-4 achieves 100% success rate in the development stage, while attaining 36% improvement on average one pass rate across alternative LLMs in the deployment stage. In both stages, DS-Agent achieves the best rank in performance, costing $1.60 and \$0.13 per run with GPT-4, respectively. Our data and code are open-sourced at https://github.com/guosyjlu/DS-Agent.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-guo24b, title = {{DS}-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning}, author = {Guo, Siyuan and Deng, Cheng and Wen, Ying and Chen, Hechang and Chang, Yi and Wang, Jun}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {16813--16848}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/guo24b/guo24b.pdf}, url = {https://proceedings.mlr.press/v235/guo24b.html}, abstract = {In this work, we investigate the potential of large language models (LLMs) based agents to automate data science tasks, with the goal of comprehending task requirements, then building and training the best-fit machine learning models. Despite their widespread success, existing LLM agents are hindered by generating unreasonable experiment plans within this scenario. To this end, we present DS-Agent, a novel automatic framework that harnesses LLM agent and case-based reasoning (CBR). In the development stage, DS-Agent follows the CBR framework to structure an automatic iteration pipeline, which can flexibly capitalize on the expert knowledge from Kaggle, and facilitate consistent performance improvement through the feedback mechanism. Moreover, DS-Agent implements a low-resource deployment stage with a simplified CBR paradigm to adapt past successful solutions from the development stage for direct code generation, significantly reducing the demand on foundational capabilities of LLMs. Empirically, DS-Agent with GPT-4 achieves 100% success rate in the development stage, while attaining 36% improvement on average one pass rate across alternative LLMs in the deployment stage. In both stages, DS-Agent achieves the best rank in performance, costing $1.60 and \$0.13 per run with GPT-4, respectively. Our data and code are open-sourced at https://github.com/guosyjlu/DS-Agent.} }
Endnote
%0 Conference Paper %T DS-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning %A Siyuan Guo %A Cheng Deng %A Ying Wen %A Hechang Chen %A Yi Chang %A Jun Wang %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-guo24b %I PMLR %P 16813--16848 %U https://proceedings.mlr.press/v235/guo24b.html %V 235 %X In this work, we investigate the potential of large language models (LLMs) based agents to automate data science tasks, with the goal of comprehending task requirements, then building and training the best-fit machine learning models. Despite their widespread success, existing LLM agents are hindered by generating unreasonable experiment plans within this scenario. To this end, we present DS-Agent, a novel automatic framework that harnesses LLM agent and case-based reasoning (CBR). In the development stage, DS-Agent follows the CBR framework to structure an automatic iteration pipeline, which can flexibly capitalize on the expert knowledge from Kaggle, and facilitate consistent performance improvement through the feedback mechanism. Moreover, DS-Agent implements a low-resource deployment stage with a simplified CBR paradigm to adapt past successful solutions from the development stage for direct code generation, significantly reducing the demand on foundational capabilities of LLMs. Empirically, DS-Agent with GPT-4 achieves 100% success rate in the development stage, while attaining 36% improvement on average one pass rate across alternative LLMs in the deployment stage. In both stages, DS-Agent achieves the best rank in performance, costing $1.60 and \$0.13 per run with GPT-4, respectively. Our data and code are open-sourced at https://github.com/guosyjlu/DS-Agent.
APA
Guo, S., Deng, C., Wen, Y., Chen, H., Chang, Y. & Wang, J.. (2024). DS-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:16813-16848 Available from https://proceedings.mlr.press/v235/guo24b.html.

Related Material