# Bayesian Optimization in AlphaGo

arXiv: https://arxiv.org/abs/1812.06855 · HTML: https://mlsanity.com/paper/1812.06855/

## Key facts

- Cited by 55 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Dec 2018.

## Built on by (most cited first, 30 of 55)

- [Well-tuned Simple Nets Excel on Tabular Datasets](https://mlsanity.com/paper/2106.11189/) (2021)
- [Tree-Structured Parzen Estimator: Understanding Its Algorithm Components and Their Roles for Better Empirical Performance](https://mlsanity.com/paper/2304.11127/) (2023)
- [Automated Reinforcement Learning (AutoRL): A Survey and Open Problems](https://mlsanity.com/paper/2201.03916/) (2022)
- [πBO: Augmenting Acquisition Functions with User Beliefs for Bayesian Optimization](https://mlsanity.com/paper/2204.11051/) (2022)
- [Joint Entropy Search for Maximally-Informed Bayesian Optimization](https://mlsanity.com/paper/2206.04771/) (2022)
- [GIBBON: General-purpose Information-Based Bayesian OptimisatioN](https://mlsanity.com/paper/2102.03324/) (2021)
- [Achieving on-Mobile Real-Time Super-Resolution with Neural Architecture and Pruning Search](https://mlsanity.com/paper/2108.08910/) (2021)
- [PriorBand: Practical Hyperparameter Optimization in the Age of Deep Learning](https://mlsanity.com/paper/2306.12370/) (2023)
- [Active Offline Policy Selection](https://mlsanity.com/paper/2106.10251/) (2021)
- [Self-Correcting Bayesian Optimization through Bayesian Active Learning](https://mlsanity.com/paper/2304.11005/) (2023)
- [Automatic Termination for Hyperparameter Optimization](https://mlsanity.com/paper/2104.08166/) (2021)
- [Bayesian Generational Population-Based Training](https://mlsanity.com/paper/2207.09405/) (2022)
- [Risk-averse Heteroscedastic Bayesian Optimization](https://mlsanity.com/paper/2111.03637/) (2021)
- [Can Fairness be Automated? Guidelines and Opportunities for Fairness-aware AutoML](https://mlsanity.com/paper/2303.08485/) (2023)
- [PED-ANOVA: Efficiently Quantifying Hyperparameter Importance in Arbitrary Subspaces](https://mlsanity.com/paper/2304.10255/) (2023)
- [Position: A Call to Action for a Human-Centered AutoML Paradigm](https://mlsanity.com/paper/2406.03348/) (2024)
- [Bayesian Optimization via Continual Variational Last Layer Training](https://mlsanity.com/paper/2412.09477/) (2024)
- [Neural Process for Black-Box Model Optimization Under Bayesian Framework](https://mlsanity.com/paper/2104.02487/) (2021)
- [Automated Reinforcement Learning: An Overview](https://mlsanity.com/paper/2201.05000/) (2022)
- [c-TPE: Tree-structured Parzen Estimator with Inequality Constraints for Expensive Hyperparameter Optimization](https://mlsanity.com/paper/2211.14411/) (2022)
- [OpenBox: A Python Toolkit for Generalized Black-box Optimization](https://mlsanity.com/paper/2304.13339/) (2023)
- [Tuning Mixed Input Hyperparameters on the Fly for Efficient Population Based AutoRL](https://mlsanity.com/paper/2106.15883/) (2021)
- [Scalable Bayesian Optimization via Focalized Sparse Gaussian Processes](https://mlsanity.com/paper/2412.20375/) (2024)
- [Automatic Mapping of the Best-Suited DNN Pruning Schemes for Real-Time Mobile Acceleration](https://mlsanity.com/paper/2111.11581/) (2021)
- [Event-Triggered Time-Varying Bayesian Optimization](https://mlsanity.com/paper/2208.10790/) (2022)
- [Speeding Up Multi-Objective Hyperparameter Optimization by Task Similarity-Based Meta-Learning for the Tree-Structured Parzen Estimator](https://mlsanity.com/paper/2212.06751/) (2022)
- [Episodic Policy Gradient Training](https://mlsanity.com/paper/2112.01853/) (2021)
- [Bayesian Optimization For Multi-Objective Mixed-Variable Problems](https://mlsanity.com/paper/2201.12767/) (2022)
- [High-dimensional Bayesian Optimization with Group Testing](https://mlsanity.com/paper/2310.03515/) (2023)
- [Mixed Variable Bayesian Optimization with Frequency Modulated Kernels](https://mlsanity.com/paper/2102.12792/) (2021)

## Abstract (arXiv)

During the development of AlphaGo, its many hyper-parameters were tuned with Bayesian optimization multiple times. This automatic tuning process resulted in substantial improvements in playing strength. For example, prior to the match with Lee Sedol, we tuned the latest AlphaGo agent and this improved its win-rate from 50% to 66.5% in self-play games. This tuned version was deployed in the final match. Of course, since we tuned AlphaGo many times during its development cycle, the compounded contribution was even higher than this percentage. It is our hope that this brief case study will be of interest to Go fans, and also provide Bayesian optimization practitioners with some insights and inspiration.