ゲーム用機械学習プログラムが、画期的な科学ツールの開発に貢献(Machine learning program for games inspires development of groundbreaking scientific tool)

ad
ad

2022-05-03 アルゴンヌ国立研究所(ANL)

私たちは、反復学習と強化学習によって新しいスキルを身につけます。試行錯誤しながら、良い結果につながる行動を繰り返し、悪い結果を避け、その中間の結果を改善しようとするのです。研究者たちは現在、強化学習を用いた人工知能の一種に基づくアルゴリズムを設計している。このアルゴリズムは、化学合成や創薬の自動化、さらにはチェスや囲碁のようなゲームのプレイにも応用されています。
米国エネルギー省(DOE)のアルゴンヌ国立研究所の科学者たちは、さらに別の用途で強化学習アルゴリズムを開発しました。それは、原子や分子のスケールで材料の特性をモデル化するためのもので、材料の発見を大幅にスピードアップさせるはずです。

<関連情報>

高次元ポテンシャルエネルギーモデル開発のための連続作用空間における学習 Learning in continuous action space for developing high dimensional potential energy models

Sukriti Manna,Troy D. Loeffler,Rohit Batra,Suvo Banik,Henry Chan,Bilvin Varughese,Kiran Sasikumar,Michael Sternberg,Tom Peterka,Mathew J. Cherukara,Stephen K. Gray,Bobby G. Sumpter & Subramanian K. R. S. Sankaranarayanan
Nature Communications  Published: 18 January 2022
DOI:https://doi.org/10.1038/s41467-021-27849-6

Abstract

Reinforcement learning (RL) approaches that combine a tree search with deep learning have found remarkable success in searching exorbitantly large, albeit discrete action spaces, as in chess, Shogi and Go. Many real-world materials discovery and design applications, however, involve multi-dimensional search problems and learning domains that have continuous action spaces. Exploring high-dimensional potential energy models of materials is an example. Traditionally, these searches are time consuming (often several years for a single bulk system) and driven by human intuition and/or expertise and more recently by global/local optimization searches that have issues with convergence and/or do not scale well with the search dimensionality. Here, in a departure from discrete action and other gradient-based approaches, we introduce a RL strategy based on decision trees that incorporates modified rewards for improved exploration, efficient sampling during playouts and a “window scaling scheme” for enhanced exploitation, to enable efficient and scalable search for continuous action space problems. Using high-dimensional artificial landscapes and control RL problems, we successfully benchmark our approach against popular global optimization schemes and state of the art policy gradient methods, respectively. We demonstrate its efficacy to parameterize potential models (physics based and high-dimensional neural networks) for 54 different elemental systems across the periodic table as well as alloys. We analyze error trends across different elements in the latent space and trace their origin to elemental structural diversity and the smoothness of the element energy surface. Broadly, our RL strategy will be applicable to many other physical science problems involving search over continuous action spaces.

コメント

タイトルとURLをコピーしました