Filters








23,714 Hits in 5.6 sec

Benchmarking Active Learning Strategies for Materials Optimization and Discovery [article]

Alex Wang, Haotong Liang, Austin McDannald, Ichiro Takeuchi, A. Gilad Kusne
2022 arXiv   pre-print
Popular active learning methods along with a recent scientific active learning method are benchmarked for their materials optimization performance.  ...  As the number, diversity, and uses for active learning strategies grow, there is an associated growing necessity for real-world reference datasets to benchmark strategies.  ...  Acknowledgement: This work was funded by NIST Cooperative Agreement 70NANB17H301 and an Office of Naval Research MURI through grant #N00014-17-1-2661.  ... 
arXiv:2204.05838v1 fatcat:fmfm4s3bnbf5patd6u6kp35p2a

Benchmarking Active Learning Strategies for Materials Optimization and Discovery

Alex Wang, Haotong Liang, Austin McDannald, Ichiro Takeuchi, A Gilad Kusne
2022 Oxford Open Materials Science  
Results Popular active learning methods along with a recent scientific active learning method are benchmarked for their materials optimization performance.  ...  As the number, diversity, and uses for active learning strategies grow, there is an associated growing necessity for real-world reference datasets to benchmark strategies.  ...  Acknowledgement: This work was funded by NIST Cooperative Agreement 70NANB17H301 and an Office of Naval Research MURI through grant #N00014-17-1-2661.  ... 
doi:10.1093/oxfmat/itac006 fatcat:vxumnp5m5vbr5ekvsfywqdccbu

Benchmarking the acceleration of materials discovery by sequential learning

Santosh S Suram, Brian Rohr, Helge S Stein, Dan Guevarra, Yu Wang, Joel A Haber, Muratahan Aykol, John Gregoire
2020 Chemical Science  
Sequential learning (SL) strategies, i.e. iteratively updating a machine learning model to guide experiments, have been proposed to significantly accelerate materials discovery and research.  ...  Applications on computational datasets and a...  ...  Acknowledgements Notes and references  ... 
doi:10.1039/c9sc05999g pmid:34084328 pmcid:PMC8157525 fatcat:am546ta46zaerd2gtb3nzzhtpa

Screening of antibacterial compounds with novel structure from the FDA approved drugs using machine learning methods

Wen-Xing Li, Xin Tong, Peng-Peng Yang, Yang Zheng, Ji-Hao Liang, Gong-Hua Li, Dahai Liu, Dao-Gang Guan, Shao-Xing Dai
2022 Aging  
The screening results showed that 1087 small-molecule drugs have potential antibacterial activity and 154 of them are FDA-approved antibacterial drugs, which accounts for 76.2% of the approved antibacterial  ...  In this study, an antibacterial compound predictor was constructed using the support vector machines and random forest methods and the data of the active and inactive antibacterial compounds from the ChEMBL  ...  G820282016 for DGG), and the National Natural Science Foundation of China (Grant Nos. 31501080 and 32070676 for DGG).  ... 
doi:10.18632/aging.203887 pmid:35150482 pmcid:PMC8876917 fatcat:7yeadvkj7ramlphexdhotkllfq

Olympus: a benchmarking framework for noisy optimization and experiment planning [article]

Florian Häse and Matteo Aldeghi and Riley J. Hickman and Loïc M. Roch and Melodie Christensen and Elena Liles and Jason E. Hein and Alán Aspuru-Guzik
2021 arXiv   pre-print
We introduce Olympus, a software package that provides a consistent and easy-to-use framework for benchmarking optimization algorithms against realistic experiments emulated via probabilistic deep-learning  ...  In chemistry and materials science, recent growth in laboratory digitization and automation has sparked interest in optimization-guided autonomous discovery and closed-loop experimentation.  ...  Several different optimization strategies have already been used for automated scientific discovery.  ... 
arXiv:2010.04153v2 fatcat:qhyayvd76ndzfafv2c3ekp7g5a

Autonomous discovery in the chemical sciences part II: Outlook

Connor W Coley, Natalie S Eyke, Klavs F. Jensen
2019 Angewandte Chemie International Edition  
experiments for validation, select experiments, and evaluate whether we are making progress towards the ultimate goal of autonomous discovery.  ...  It is increasingly important to articulate what the role of automation and computation has been in the scientific process and how that has or has not accelerated discovery.  ...  This work was supported by the Machine Learning for Pharmaceutical Discovery and Synthesis Consortium and the DARPA Make-It program under contract ARO W911NF-16-2-0023.  ... 
doi:10.1002/anie.201909989 pmid:31553509 fatcat:qwc5y66npbf7jjf5lq53tw5igi

Gryffin: An algorithm for Bayesian optimization of categorical variables informed by expert knowledge [article]

Florian Häse, Matteo Aldeghi, Riley J. Hickman, Loïc M. Roch, Alán Aspuru-Guzik
2021 arXiv   pre-print
In addition to comprehensive benchmarks, we demonstrate the capabilities and performance of Gryffin on three examples in materials science and chemistry: (i) the discovery of non-fullerene acceptors for  ...  Leveraging domain knowledge in the form of physicochemical descriptors, Gryffin can significantly accelerate the search for promising molecules and materials.  ...  discovery, as well as their benefits over conventional experimentation strategies, are being actively explored. 27 For example, autonomous platforms have been applied to the optimization of reaction  ... 
arXiv:2003.12127v2 fatcat:stdbhzymg5fljkckkjd3gtqw4a

Practical considerations for active machine learning in drug discovery

Daniel Reker
2020 Drug Discovery Today : Technologies  
By focusing on these practical aspects of active learning, this review aims at providing insights for scientists planning to implement active learning workflows in their discovery pipelines.  ...  Although a long established theoretical concept and introduced to drug discovery approximately 15 years ago, the deployment of active learning technology in the discovery pipelines across academia and  ...  Acknowledgements Daniel Reker is supported by the Swiss National Science Foundation (grants P2EZP3_168827 and P300P2_177833), the MIT-IBM Watson AI Lab, and the MIT SenseTime alliance.  ... 
doi:10.1016/j.ddtec.2020.06.001 pmid:33386097 fatcat:sd6rivhyc5gj3puykka2wpvixy

On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning [article]

A. Gilad Kusne, Heshan Yu, Changming Wu, Huairuo Zhang, Jason Hattrick-Simpers, Brian DeCost, Suchismita Sarker, Corey Oses, Cormac Toher, Stefano Curtarolo, Albert V. Davydov, Ritesh Agarwal (+4 others)
2020 arXiv   pre-print
Active learning - the field of machine learning (ML) dedicated to optimal experiment design, has played a part in science as far back as the 18th century when Laplace used it to guide his discovery of  ...  and property optimization, with each cycle taking seconds to minutes, resulting in the discovery of a novel epitaxial nanocomposite phase-change memory material.  ...  Active Learning -Materials Optimization: Benchmark System The target optimization for the benchmark system is maximizing remnant magnetization.  ... 
arXiv:2006.06141v2 fatcat:peddokaeifamld42ojosrc6jyi

On-the-fly closed-loop materials discovery via Bayesian active learning

A. Gilad Kusne, Heshan Yu, Changming Wu, Huairuo Zhang, Jason Hattrick-Simpers, Brian DeCost, Suchismita Sarker, Corey Oses, Cormac Toher, Stefano Curtarolo, Albert V. Davydov, Ritesh Agarwal (+4 others)
2020 Nature Communications  
We demonstrate an autonomous materials discovery methodology for functional inorganic compounds which allow scientists to fail smarter, learn faster, and spend less resources in their studies, while simultaneously  ...  The real-time closed-loop, autonomous system for materials exploration and optimization (CAMEO) is implemented at the synchrotron beamline to accelerate the interconnected tasks of phase mapping and property  ...  Acknowledgements We acknowledge Xiaohang Zhang for assistance with thin film characterization and acknowledges support from Materials Genome Initiative funding allocated to NIST.  ... 
doi:10.1038/s41467-020-19597-w pmid:33235197 fatcat:hzmuj3yxungdblpzpe22y6pk4u

Exploring Deep Recurrent Models with Reinforcement Learning for Molecule Design

Daniel Neil, Marwin H. S. Segler, Laura Guasch, Mohamed Ahmed, Dean Plumbley, Matthew Sellwood, Nathan Brown
2018 International Conference on Learning Representations  
However significant challenges yet remain for computational methods, despite recent advances such as deep recurrent networks and reinforcement learning strategies for sequence generation, and it can be  ...  This work proposes 19 benchmarks selected by subject experts, expands smaller datasets previously used to approximately 1.1 million training molecules, and explores how to apply new reinforcement learning  ...  RL is a natural environment for drug discovery, which requires online learning balanced against expensive sample evaluation and generation.  ... 
dblp:conf/iclr/NeilSGAPSB18 fatcat:kauxjsiu4fgltlkz763onsupgu

Gemini: Dynamic Bias Correction for Autonomous Experimentation and Molecular Simulation [article]

Riley J. Hickman, Florian Häse, Loïc M. Roch, Alán Aspuru-Guzik
2021 arXiv   pre-print
Finally, we simulate an autonomous materials discovery platform for optimizing the activity of electrocatalysts for the oxygen evolution reaction.  ...  Bayesian optimization has emerged as a powerful strategy to accelerate scientific discovery by means of autonomous experimentation.  ...  Finally, we simulate an autonomous materials discovery platform for optimizing the activity of electrocatalysts for the oxygen evolution reaction.  ... 
arXiv:2103.03391v1 fatcat:n2e2h7e34vacvhw4ovti4afz6q

Sequential learning to accelerate discovery of alkali-activated binders

Christoph Völker, Rafia Firdous, Dietmar Stephan, Sabine Kruschwitz
2021 Journal of Materials Science  
The SL approach combines machine learning models and feedback from real experiments. For this purpose, 131 data points were collected from different publications.  ...  However, as yet there are no sufficiently accurate material models to effectively predict the AAB properties, thus making optimal mix design highly costly and reducing the attractiveness of such binders  ...  Acknowledgements We would like to thank Philipp Benner from the Federal Institute for Materials Research and Testing for the many valuable discussions on the background of the mathematical methods and  ... 
doi:10.1007/s10853-021-06324-z fatcat:c3ixtxn4jze3nb5v6lxtn3xppm

Assisting Multitargeted Ligand Affinity Prediction of Receptor Tyrosine Kinases Associated Nonsmall Cell Lung Cancer Treatment with Multitasking Principal Neighborhood Aggregation

Fahsai Nakarin, Kajjana Boonpalit, Jiramet Kinchagawat, Patcharapol Wachiraphan, Thanyada Rungrotmongkol, Sarana Nutanong
2022 Molecules  
In this investigation, we employed machine learning models to provide a computationally affordable means for computer-aided screening to accelerate the discovery of potential drug compounds.  ...  A multitargeted therapeutic approach with hybrid drugs is a promising strategy to enhance anticancer efficiency and overcome drug resistance in nonsmall cell lung cancer (NSCLC) treatment.  ...  The GBRT benchmark model was provided by Jaruratana Khantiratana for both optimization and training processes. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/molecules27041226 pmid:35209011 pmcid:PMC8878292 fatcat:dwlbzz5qj5bspdlvquzyvj6cdi

Towards automated design of corrosion resistant alloy coatings with an autonomous scanning droplet cell [article]

Brian DeCost, Howie Joress, Suchismita Sarker, Apurva Mehta, Jason Hattrick-Simpers
2022 arXiv   pre-print
Automation and machine learning are currently driving rapid innovation in high throughput and autonomous materials design and discovery.  ...  This emerging research area presents new opportunities to accelerate materials synthesis, evaluation, and hence discovery and design.  ...  BD and HJ acknowledge partial support from the National Research Council Research Associate Program.  ... 
arXiv:2203.17049v1 fatcat:jp5td2shcbdwrdpdep4ia5lzqa
« Previous Showing results 1 — 15 out of 23,714 results