Open Source Vizier: Distributed Infrastructure and API for Reliable and Flexible Blackbox Optimization

Xingyou Song, Sagi Perel, Chansoo Lee, Greg Kochanski, Daniel Golovin

AutoML 2022 pp. 8/1-17

doi:10.48550/arXiv.2207.13676 /automl/2022/song2022automl-open/

Abstract

Vizier is the de-facto blackbox optimization service across Google, having optimized some of Google’s largest products and research efforts. To operate at the scale of tuning thousands of users’ critical systems, Vizier solved key design challenges in providing multiple different features, while remaining fully fault-tolerant. In this paper, we introduce Open Source (OSS) Vizier, a Python-based interface for blackbox optimization and research, based on the Google-internal Vizier infrastructure and framework. OSS Vizier provides an API capable of defining and solving a wide variety of optimization problems, including multi-metric, early stopping, transfer learning, and conditional search. Furthermore, it is designed to be a distributed system that assures reliability, and allows multiple parallel evaluations of the user’s objective function. The flexible RPC-based infrastructure allows users to access OSS Vizier from binaries written in any language. OSS Vizier also provides a back-end (”Pythia”) API that gives algorithm authors a way to interface new algorithms with the core Vizier system. OSS Vizier is available at \url{https://github.com/google/vizier}.

PDF AutoML OpenReview Semantic Scholar

Cite

Text

Song et al. "Open Source Vizier: Distributed Infrastructure and API for Reliable and Flexible Blackbox Optimization." Proceedings of the First International Conference on Automated Machine Learning, 2022. doi:10.48550/arXiv.2207.13676

Markdown

[Song et al. "Open Source Vizier: Distributed Infrastructure and API for Reliable and Flexible Blackbox Optimization." Proceedings of the First International Conference on Automated Machine Learning, 2022.](https://mlanthology.org/automl/2022/song2022automl-open/) doi:10.48550/arXiv.2207.13676

BibTeX

@inproceedings{song2022automl-open,
  title     = {{Open Source Vizier: Distributed Infrastructure and API for Reliable and Flexible Blackbox Optimization}},
  author    = {Song, Xingyou and Perel, Sagi and Lee, Chansoo and Kochanski, Greg and Golovin, Daniel},
  booktitle = {Proceedings of the First International Conference on Automated Machine Learning},
  year      = {2022},
  pages     = {8/1-17},
  doi       = {10.48550/arXiv.2207.13676},
  volume    = {188},
  url       = {https://mlanthology.org/automl/2022/song2022automl-open/}
}