Deep Clustering based Boundary-Decoder Net for Inter and Intra Layer Stress Prediction of Heterogeneous Integrated IC Chip
arXiv:2602.21601v1 Announce Type: new Abstract: High stress occurs when 3D heterogeneous IC packages are subjected to thermal cycling at extreme temperatures. Stress mainly occurs at the interface between different materials. We investigate stress image using latent space representation which is based on using deep generative model (DGM). However, most DGM approaches are unsupervised, meaning they resort to image pairing (input and output) to train DGM. Instead, we rely on a recent boundary-decoder (BD) net, which uses boundary condition and image pairing for stress modeling. The boundary net maps material parameters to the latent space co-shared by its image counterpart. Because such a setup is dimensionally wise ill-posed, we further couple BD net with deep clustering. To access the performance of our proposed method, we simulate an IC chip dataset comprising of 1825 stress images. We compare our new approach using variants of BD net as well as a baseline approach. We show that our
arXiv:2602.21601v1 Announce Type: new Abstract: High stress occurs when 3D heterogeneous IC packages are subjected to thermal cycling at extreme temperatures. Stress mainly occurs at the interface between different materials. We investigate stress image using latent space representation which is based on using deep generative model (DGM). However, most DGM approaches are unsupervised, meaning they resort to image pairing (input and output) to train DGM. Instead, we rely on a recent boundary-decoder (BD) net, which uses boundary condition and image pairing for stress modeling. The boundary net maps material parameters to the latent space co-shared by its image counterpart. Because such a setup is dimensionally wise ill-posed, we further couple BD net with deep clustering. To access the performance of our proposed method, we simulate an IC chip dataset comprising of 1825 stress images. We compare our new approach using variants of BD net as well as a baseline approach. We show that our approach is able to outperform all the comparison in terms of train and test error reduction.
Executive Summary
This article proposes a novel approach for predicting inter and intra layer stress in heterogeneous integrated IC chips using a deep clustering based boundary-decoder net. The method leverages a deep generative model and boundary conditions to map material parameters to a latent space, allowing for accurate stress modeling. The approach is evaluated on a dataset of 1825 stress images and demonstrates improved performance compared to baseline methods. The proposed method has significant implications for the design and development of reliable and efficient IC chips, particularly in applications where thermal stress is a critical concern.
Key Points
- ▸ Deep clustering based boundary-decoder net for stress prediction
- ▸ Use of latent space representation for stress modeling
- ▸ Evaluation on a dataset of 1825 stress images
Merits
Improved Accuracy
The proposed method demonstrates improved performance compared to baseline methods, with reduced train and test errors.
Demerits
Dimensional Complexity
The setup of the boundary-decoder net is dimensionally ill-posed, requiring the use of deep clustering to mitigate this issue.
Expert Commentary
The proposed method represents a significant advancement in the field of IC chip design and development, particularly in the context of thermal stress prediction. The use of deep clustering and boundary-decoder nets offers a powerful approach for modeling complex stress phenomena. However, further research is needed to fully explore the potential of this method and to address the dimensional complexity of the setup. The implications of this work are far-reaching, with potential applications in a wide range of fields, from consumer electronics to aerospace engineering.
Recommendations
- ✓ Further evaluation of the proposed method on larger and more diverse datasets
- ✓ Exploration of the potential applications of this method in other fields, such as materials science and mechanical engineering
