New paper - Model Construction from Experiments

We have a new paper out today on unbiased construction of constitutive relations for soft materials from experiments via rheology-informed neural networks. You can read a press release on the paper here.

Traditional methods often rely on simplifying assumptions, leading to biases and limitations in precise material-specific characterization. In contrast, the proposed rheology-informed neural network framework learns the hidden rheology of complex fluids from a limited number of experiments, enabling the construction of unbiased material-specific constitutive relations. This approach accurately describes a wide range of bulk dynamical behaviors of soft materials while providing insights into their underlying physics. Overall, the method offers an efficient and insightful means of constructing constitutive models, with implications beyond complex fluids and into other disciplines.

Previous post
Soohee Bae Defends Master's Thesis Next post
Donya Dabiri's Oral Exam