Cooperative neural networks cut design data cost for automotive parts
Stream-Based Active Learning with Cooperative Neural Networks for Data-Efficient Partial Inverse Design: An Automotive Glass Run Channel Case Study
Machine LearningArtificial Intelligence
Summary
Designing parts in engineering usually needs lots of expensive tests or simulations to figure out the right settings. The authors made a smarter system that picks only the most useful design examples to actually test, saving a lot of time and money. They focus on problems where some design parts are known but others are guessed to meet a goal. Their approach works well on a big dataset of simulated car window channel designs, using far fewer test examples than usual but still making accurate predictions.
What this means in practice
- •For automotive design engineers: Reduce costly simulations by selecting only the most informative designs for labeling in automotive part inverse design problems.
- •For industrial design data scientists: Improve data efficiency in design optimization by integrating uncertainty-based active learning with cooperative neural network models.
Authors
Agung Nugraha, Hyerin Kwon, Heungjun Im, Gian Antariksa, Jihwan Lee
Abstract
Inverse design in engineering often runs into a simple problem. Each labeled training sample must be produced through expensive simulation, so building a large dataset is slow and costly. This study addresses that problem for partial inverse design, where only some design variables are specified and the rest must be inferred to reach a target performance value. We propose CoNN-AL, a framework for data-efficient partial inverse design that adds stream-based active learning to the Cooperative Neural Network with Denoising Autoencoder (CoNN-DAE). The model estimates predictive uncertainty through Monte Carlo dropout and uses it to decide, in real time, which incoming candidate samples are worth labeling, so the limited labeling budget is spent on the most informative designs. We validate the framework on a real-world automotive glass run channel dataset of more than 900,000 unique simulated designs. With only 20,000 actively selected labels, about 2.3% of the training pool, CoNN-AL reaches R-squared values of 0.967 to 0.982 across all missing-variable levels, approaching the upper-bound models trained on far more data. It reaches R-squared of at least 0.95 with 30 to 40% fewer labels than random sampling at the more difficult missing-variable levels and, at the most challenging level, is the only strategy in this study to reach R-squared of 0.98. Together with this work, we publicly release the dataset to support future research on data-driven design.