Efficient construction method for phase diagrams using uncertainty sampling

Terayama, Kei; Tamura, Ryo; Nosé, Yukihiko; Hiramatsu, Hidenori; Hosono, Hideo; Okuno, Yasushi; Tsuda, Koji

doi:10.1103/physrevmaterials.3.033802

Cited by 34 publications

(39 citation statements)

References 24 publications

(21 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we examined phase diagrams of parameter spaces with two phases, successful and failed. The phase diagram construction method based on US is also applicable in cases where there are more than two types of phases, as described in [22]. For example, in the case of the F1-motor system, because the substep (90 + 30) for 120-degree rotation was frequently observed at lower ATP concentrations in in-vitro experiments [35] there are three possibilities: 120-degree rotation at once (without substep), substep rotation (90 + 30 degree), and others (failure for rotation).…”

Section: Discussionmentioning

confidence: 99%

“…An active learning-based approach proposed by some authors for efficiently constructing a phase diagram [22] is expected to be useful for identifying regions of successful parameters. The approach adopted uncertainty sampling, which is a type of active learning, for preferentially sampling around predicted boundaries of multiple discrete classes by machine learning to quickly determine the boundaries.…”

Section: Parameter Sampling By Usmentioning

confidence: 99%

“…Figure 1b shows the procedure of successful region search based on US. We implemented the method based on a combination of the label propagation (LP) [28] and the least confidence method [29], which showed the best performance among algorithms described in previous studies [22,23]. We assigned a successful or failed label to each parameter based on a success rate.…”

Section: Parameter Sampling By Usmentioning

confidence: 99%

“…That is, if the predicted probabilities of success and failure are almost the same, the score is relatively high. We used default hyperparameters in the previous paper [22]. It was shown that US works well under default parameters.…”

Section: Parameter Sampling By Usmentioning

confidence: 99%

“…However, it is not necessarily appropriate to efficiently search for parameters beyond a certain criterion. In contrast, one of the authors recently proposed an effective sampling method [22,23] for constructing phase diagrams based on uncertainty sampling (US), a type of active learning technique. The method based on US can efficiently determine phases to examine the phase boundary preferentially when two or more phases are sampled.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Exploring Successful Parameter Region for Coarse-Grained Simulation of Biomolecules by Bayesian Optimization and Active Learning

Kanada¹,

Tokuhisa

Tsuda

et al. 2020

Biomolecules

Self Cite

View full text Add to dashboard Cite

Accompanied with an increase of revealed biomolecular structures owing to advancements in structural biology, the molecular dynamics (MD) approach, especially coarse-grained (CG) MD suitable for macromolecules, is becoming increasingly important for elucidating their dynamics and behavior. In fact, CG-MD simulation has succeeded in qualitatively reproducing numerous biological processes for various biomolecules such as conformational changes and protein folding with reasonable calculation costs. However, CG-MD simulations strongly depend on various parameters, and selecting an appropriate parameter set is necessary to reproduce a particular biological process. Because exhaustive examination of all candidate parameters is inefficient, it is important to identify successful parameters. Furthermore, the successful region, in which the desired process is reproducible, is essential for describing the detailed mechanics of functional processes and environmental sensitivity and robustness. We propose an efficient search method for identifying the successful region by using two machine learning techniques, Bayesian optimization and active learning. We evaluated its performance using F1-ATPase, a biological rotary motor, with CG-MD simulations. We successfully identified the successful region with lower computational costs (12.3% in the best case) without sacrificing accuracy compared to exhaustive search. This method can accelerate not only parameter search but also biological discussion of the detailed mechanics of functional processes and environmental sensitivity based on MD simulation studies.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Parameter Sampling By Usmentioning

confidence: 99%

Section: Parameter Sampling By Usmentioning

confidence: 99%

Section: Parameter Sampling By Usmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Exploring Successful Parameter Region for Coarse-Grained Simulation of Biomolecules by Bayesian Optimization and Active Learning

Kanada¹,

Tokuhisa

Tsuda

et al. 2020

Biomolecules

Self Cite

View full text Add to dashboard Cite

show abstract

Global‐Oriented Strategy for Searching Ultrastrength Martensitic Stainless Steels

Zhao

et al. 2022

Advcd Theory and Sims

View full text Add to dashboard Cite

An intelligent alloy design strategy integrating machine learning and adaptive sampling is proposed and successfully applied to an example of martensitic stainless steel. Short iterative experiments prove the feasibility of this strategy in quickly finding new alloys with hardness higher than that of the current best and reasonably augmenting training data to create a high-confidence prediction model. A credible relationship between the composition and hardness is demonstrated by the proposed model and the most promising candidate in the designed space is identified. In contrast to the traditional approaches, this strategy can meet the goal of global quest as it offers the advantages of flexibility, reliability, and efficiency. The suggested strategy can be extended to guide the experimental design of other materials.

show abstract

Bayesian Optimization‐guided Discovery of High‐performance Methane Combustion Catalysts based on Multi‐component PtPd@CeZrOx Core–Shell Nanospheres

Feng,

Gong,

Liu

et al. 2023

Angewandte Chemie

View full text Add to dashboard Cite

Formula regulation of multi‐component catalysts by manual search is undoubtedly a time‐consuming task, which has severely impeded the development efficiency of high‐performance catalysts. In this work, PtPd@CeZrOx core‐shell nanospheres, as a successful case study, is explicitly demonstrated how Bayesian optimization (BO) accelerates the discovery of methane combustion catalysts with the optimal formula ratio, which directly results in a lower temperature of methane conversion than ones reported hitherto in literature. As a consequence, the best sample obtained in the present work could be efficiently developed after two rounds of iterations, containing only 18 controllable experiments in all that is far less than the common human workload via the traditional trial‐and‐error search for optimal compositions. Further, this BO‐based machine learning strategy can be straightforward extended to serve the autonomous discovery in multi‐component material systems for other desired properties, showing promising opportunities to practical applications in future.

show abstract

Efficient construction method for phase diagrams using uncertainty sampling

Cited by 34 publications

References 24 publications

Exploring Successful Parameter Region for Coarse-Grained Simulation of Biomolecules by Bayesian Optimization and Active Learning

Exploring Successful Parameter Region for Coarse-Grained Simulation of Biomolecules by Bayesian Optimization and Active Learning

Global‐Oriented Strategy for Searching Ultrastrength Martensitic Stainless Steels

Bayesian Optimization‐guided Discovery of High‐performance Methane Combustion Catalysts based on Multi‐component PtPd@CeZrOx Core–Shell Nanospheres

Contact Info

Product

Resources

About