Speaker
Description
As computational strength keeps growing, deep learning has emerged as a powerful technique for addressing complex tasks and solving problems with intricate logic. Researchers are starting to leverage deep learning methods to tackle all kinds of challenges, including inversion problems in different materials. However, training deep neural networks (DNNs) for such tasks requires extensive datasets, which may not always be available. In this work, we aim to utilize stress-strain curves to predict the microstructural features of the materials, but traditional DNN models alone demand substantial amounts of training data. To overcome the limitations of small datasets, we propose incorporating mechanical relationships as additional features within the model. By integrating domain-specific mechanical knowledge, our approach enables the DNN to learn effectively from limited data, enhancing feature extraction and prediction efficiency. This combined framework demonstrates how blending deep learning with physics-based constraints can improve performance and accelerate computations in data-scarce environments.
| Country | United States |
|---|---|
| Student Awards | I would like to submit this presentation into both awards |
| Acceptance of the Terms & Conditions | Click here to agree |
