[{"id":987,"title":"Democratizing experimental benchmarking of materials discovery campaigns through a low-cost, self-driving optics demo","sha":"12372397dbaf594ca372f17ebbb8c2a3","user_id":null,"state":"published","body":"Self-driving labs are the future; however, the capital and expertise required can be daunting. We introduce the idea of a constrained, high-dimensional, multi-objective optimization task for less than $100, a square foot of desk space, and an hour of total setup time from the shopping cart to the first \"autonomous drive.\" We use optics rather than chemistry for our demo; after all, light is easier to move than matter. While not strictly materials-based, importantly, several core principles of a self-driving materials discovery lab are retained in this cross-domain example: sending commands to hardware to adjust physical parameters, receiving measured objective properties, decision-making via active learning, and utilizing cloud-based simulations. The demo is accessible, extensible, modular, and repeatable, making it an ideal candidate for both low-cost experimental adaptive design benchmarking and learning the principles of self-driving laboratories in a low-risk setting.","created_at":"2022-07-02T21:02:16.649Z","updated_at":"2022-09-16T23:13:58.917Z","zenodo_id":6970227,"doi":"https://doi.org/10.5281/zenodo.6970227","subject":null,"tags":["self-driving laboratory","bayesian optimization","autonomous research","demonstration","pedagogy"],"vote_count":1,"view_count":1,"score":1.1,"deleted":false,"muted":false,"tweeted":true,"attachment_file_name":"abstract.png","attachment_content_type":"image/png","attachment_file_size":228822,"attachment_updated_at":"2022-07-02T21:02:16.628Z"},{"id":1017,"title":"CLSLabs-Liquid extends the minimal working example of self-driving laboratories to liquid handling and optical spectroscopy","sha":"396816b186ddb4d5237ce41cdb65b364","user_id":null,"state":"published","body":"Materials acceleration platforms have the potential to reduce the cost- and time-to-market of new materials from 20 million USD and 20 years to as little as 1 million USD and 1 year; however, the capital and expertise are often significant. Previously, we introduced the idea of a minimal working example for a self-driving laboratory that uses RGB light-emitting diodes and a discrete-channel spectrophotometer in conjunction with optimization algorithms to match target spectra. We extend this idea to a color-matching materials optimization problem using dilute colored dyes. This optimization task costs less than 300 USD, requires less than three square feet of desk space, and less than three hours of total setup time from the shopping cart to the first “autonomous drive.” The demo is modular and extensible; additional peristaltic pump channels can be added, and the dye reservoirs can be increased. The demo was also designed so inexpensive, chemically-susceptible parts can be replaced with more expensive, chemically-resistant equivalents, and the food dye can be replaced with application-specific chemicals. This demo is a stepping stone towards more extensive, high-impact studies; for example, formulations of dozens of battery or fuel cell liquid electrolytes optimized for ionic conductivity properties.","created_at":"2022-10-29T19:16:33.508Z","updated_at":"2022-11-28T06:00:14.809Z","zenodo_id":7369997,"doi":"https://doi.org/10.5281/zenodo.7369997","subject":null,"tags":["self-driving laboratory","bayesian optimization","autonomous research","demonstration","pedagogy"],"vote_count":0,"view_count":1,"score":0.1,"deleted":false,"muted":false,"tweeted":false,"attachment_file_name":"cls-labs-liquid-summary.png","attachment_content_type":"image/png","attachment_file_size":150746,"attachment_updated_at":"2022-10-29T19:16:33.489Z"},{"id":1042,"title":"Training a large language model to do scientific data extraction for a corpus of materials properties","sha":"aabe5c12c722f2a89ed2cca1a8b5435e","user_id":null,"state":"published","body":"A wealth of experiments and simulation results are contained in the scientific literature. Extracting it in a machine-readable format is daunting as it typically requires domain expertise, rigid schema, or intensive labor. Recently, Dunn et al. (DOI: 10.48550/arXiv.2212.05238) fine-tuned a large language model, GPT-3 by OpenAI, on a collection of text snippets containing relevant materials information and structured output with surprisingly good results on held-out text examples. Here, we extend Dunn et al.'s work for use with the Materials Platform for Data Science (MPDS), a platform based on the Pauling File representing over 1000 full-time academic manpower years' worth of curated, manual data extraction. Additionally, MPDS contains approximately 200 unique types of properties covering a broad range of topics for electronic, mechanical, optical, thermodynamic, and magnetic properties. We propose fine-tuning GPT-3 using the structured information MPDS provides to merge the semantic prowess of one of the world's most significant language models to date with one of the largest experimental materials properties databases currently available. We plan to look at in- and out-of-distribution extraction from scientific texts as a function of property type and journal. This helps pave the way toward human-like semantic parsing by leveraging the vastness of the scientific literature.","created_at":"2023-01-21T00:09:09.053Z","updated_at":"2023-03-29T01:45:33.886Z","zenodo_id":7556226,"doi":"https://doi.org/10.5281/zenodo.7556226","subject":null,"tags":["machine learning; data science; llm; nlp; materials science; materials informatics"],"vote_count":2,"view_count":1,"score":2.1,"deleted":false,"muted":false,"tweeted":true,"attachment_file_name":"abstract.png","attachment_content_type":"image/png","attachment_file_size":39203,"attachment_updated_at":"2023-01-21T00:09:07.158Z"}]