Solid electrolytes for Li-ion batteries via machine learning

Computational approach inspired advancements of solid-state electrolytes for lithium secondary batteries: from first-principles to machine learning

“…9(a)). 331,332 They collected training data from experimental reports of 160 samples and selected 50 material features to avoid overfitting and ensure training accuracy. The ML model achieved high accuracy during cross-validation, with a coefficient of determination (R 2 ) of 0.97, a root mean squared error (RMSE) of 0.005, and a mean absolute error (MAE) of 0.007.…”

A review of polymers in sulfide-based hybrid solid-state electrolytes for all-solid-state lithium batteries

“…229 In addition, the number of parameters should be considered simultaneously, making it challenging to derive significant correlations from the combinations of large dataset parameters. 228,[230][231][232][233] Thus, we must use more sophisticated engineering approaches to tackle these issues. These time-consuming trial-and-error processes could be significantly streamlined by increasing adeptness by adopting computational techniques in analyzing complex data sets with many variables.…”

Computational screening of sodium solid electrolytes through unsupervised learning

“…1 However, concerns regarding a stable supply, mainly driven by the locally distributed lithium resources, underscore the pressing need to explore and develop a commercially viable Na-ion based SSE. 1,2 Nonetheless, the major drawback of Na-ion based SSE is their low ionic conductivity at room temperature (RT).…”

“…For this, machine learning (ML) techniques can be exploited for uncovering latent data structures and exceptional accuracy in prediction tasks. 2,17,[26][27][28][29][18][19][20][21][22][23][24][25] However, their application in material discovery may produce unreliable outcomes, mainly due to model overfitting from insufficient data. Our comprehensive literature survey reveals that, to date, only 34 Na-ion SSEs have experimental records of ionic conductivity.…”

“…Our comprehensive literature survey reveals that, to date, only 34 Na-ion SSEs have experimental records of ionic conductivity. While there exists a study that attempts to predict the conductivity of Na-ion SSEs using a random forest regression model trained on solid-state materials information 2 , it is worth noting that, in many instances, the same concerns regarding model overfitting are shared with previous studies aimed at uncovering Li-ion superionic SSEs.…”