Fig. 3: Predicting compressive strengths and electrical resistances of conductive aerogels.
From: Machine intelligence accelerated design of conductive MXene aerogels with programmable properties
a Comparison between the actual stress–strain curves of conductive aerogels (recipes #1–#8) and the model-predicted \({\sigma }_{30}\) values. b Comparison between the actual initial electrical resistances of conductive aerogels (recipes #1–#8) and model-predicted \({R}_{0}\) values. c By inputting specific design requests, the champion model was able to automate the inverse design processes of conductive aerogels by directly suggesting suitable sets of fabrication parameters, without the need for iterative optimization experiments. Inset shows the SEM images of two model-suggested conductive aerogels. d Comparison between actual and model-predicted \({\sigma }_{30}\) (left) and \({R}_{0}\) (right) values of conductive aerogels (recipes #9–#14). Data are presented as mean ± s.d., n = 3, with each independent experiment marked by a black or blue dot. e Violin plots of achievable \({\sigma }_{30}\) and \({R}_{0}\) values of conductive aerogels. The embedded box plot within each violin plot indicates the 25th and 75th percentiles with the median represented by the center line. Whiskers extend to 1.5 ×IQR from the box, n = 491,131. Error bars represent s.d.
