Hyper-Parameter Sweep on AlphaZero General

2019-03-19 · Hui Wang, Michael Emmerich, Mike Preuss, Aske Plaat

Since AlphaGo and AlphaGo Zero have achieved breakground successes in the game of Go, the programs have been generalized to solve other tasks. Subsequently, AlphaZero was developed to play Go, Chess and Shogi. In the literature, the algorithms are explained well. However, AlphaZero contains many parameters, and for neither AlphaGo, AlphaGo Zero nor AlphaZero, there is sufficient discussion about how to set parameter values in these algorithms. Therefore, in this paper, we choose 12 parameters in AlphaZero and evaluate how these parameters contribute to training. We focus on three objectives~(training loss, time cost and playing strength). For each parameter, we train 3 models using 3 different values~(minimum value, default value, maximum value). We use the game of play 6$\times$6 Othello, on the AlphaZeroGeneral open source re-implementation of AlphaZero. Overall, experimental results show that different values can lead to different training results, proving the importance of such a parameter sweep. We categorize these 12 parameters into time-sensitive parameters and time-friendly parameters. Moreover, through multi-objective analysis, this paper provides an insightful basis for further hyper-parameter optimization.

📄 PDF Abstract BibTeX arXiv:1903.08129

Code (1)

QueensGambit/CrazyAra mxnet

Tasks

Game of Go

Methods 이 논문이 사용한 방법론

AlphaZero AlphaZero is a reinforcement learning agent for playing board games such as Go, chess, and shogi.

Domain Knowledge in Exploration Noise in AlphaZero

2021-01-01 · Eric Weiner, George D Montañez, Aaron Trujillo, Abtin Molavi

The AlphaZero algorithm has achieved remarkable success in a variety of sequential, perfect information games including Go, Shogi and Chess. In the original paper the only hyperparameter that is changed from game to game…

Accelerating and Improving AlphaZero Using Population Based Training

2020-03-13 · Ti-Rong Wu, Ting-Han Wei, I-Chen Wu

AlphaZero has been very successful in many games. Unfortunately, it still consumes a huge amount of computing resources, the majority of which is spent in self-play. Hyperparameter tuning exacerbates the training cost si…

Hyperbolic Minesweeper is in P

2020-02-21 · Eryk Kopczyński

We show that, while Minesweeper is NP-complete, its hyperbolic variant is in P. Our proof does not rely on the rules of Minesweeper, but is valid for any puzzle based on satisfying local constraints on a graph embedded i…

valid

Minigo: A Case Study in Reproducing Reinforcement Learning Research

2019-03-06 · ICLR Workshop RML 2019 5 · Anonymous

The reproducibility of reinforcement-learning research has been highlighted as a key challenge area in the field. In this paper, we present a case study in reproducing the results of one groundbreaking algorithm, AlphaZe…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Configuration Over Selection: Hyperparameter Sensitivity Exceeds Model Differences in Open-Source LLMs for RTL Generation

2026-04-18 · Minghao Shao, Zeng Wang, Weimin Fu, Xiaolong Guo 외 arxiv

Benchmarking of open-source LLMs for hardware design focuses on which LLMs to use, while treating inference-time decoding configuration as a secondary concern. This work shows that it matters more how an LLM is configure…