Constraining Generative Models for Engineering Design with Negative Data
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917853117022208 |
|---|---|
| author | Regenwetter, Lyle Giannone, Giorgio Srivastava, Akash Gutfreund, Dan Ahmed, Faez |
| author_facet | Regenwetter, Lyle Giannone, Giorgio Srivastava, Akash Gutfreund, Dan Ahmed, Faez |
| contents | Generative models have recently achieved remarkable success and widespread adoption in society, yet they often struggle to generate realistic and accurate outputs. This challenge extends beyond language and vision into fields like engineering design, where safety-critical engineering standards and non-negotiable physical laws tightly constrain what outputs are considered acceptable. In this work, we introduce a novel training method to guide a generative model toward constraint-satisfying outputs using `negative data' -- examples of what to avoid. Our negative-data generative model (NDGM) formulation easily outperforms classic models, generating 1/6 as many constraint-violating samples using 1/8 as much data in certain problems. It also consistently outperforms other baselines, achieving a balance between constraint satisfaction and distributional similarity that is unsurpassed by any other model in 12 of the 14 problems tested. This widespread superiority is rigorously demonstrated across numerous synthetic tests and real engineering problems, such as ship hull synthesis with hydrodynamic constraints and vehicle design with impact safety constraints. Our benchmarks showcase both the best-in-class performance of our new NDGM formulation and the overall dominance of NDGMs versus classic generative models. We publicly release the code and benchmarks at https://github.com/Lyleregenwetter/NDGMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2306_15166 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Constraining Generative Models for Engineering Design with Negative Data Regenwetter, Lyle Giannone, Giorgio Srivastava, Akash Gutfreund, Dan Ahmed, Faez Machine Learning Computational Engineering, Finance, and Science Generative models have recently achieved remarkable success and widespread adoption in society, yet they often struggle to generate realistic and accurate outputs. This challenge extends beyond language and vision into fields like engineering design, where safety-critical engineering standards and non-negotiable physical laws tightly constrain what outputs are considered acceptable. In this work, we introduce a novel training method to guide a generative model toward constraint-satisfying outputs using `negative data' -- examples of what to avoid. Our negative-data generative model (NDGM) formulation easily outperforms classic models, generating 1/6 as many constraint-violating samples using 1/8 as much data in certain problems. It also consistently outperforms other baselines, achieving a balance between constraint satisfaction and distributional similarity that is unsurpassed by any other model in 12 of the 14 problems tested. This widespread superiority is rigorously demonstrated across numerous synthetic tests and real engineering problems, such as ship hull synthesis with hydrodynamic constraints and vehicle design with impact safety constraints. Our benchmarks showcase both the best-in-class performance of our new NDGM formulation and the overall dominance of NDGMs versus classic generative models. We publicly release the code and benchmarks at https://github.com/Lyleregenwetter/NDGMs. |
| title | Constraining Generative Models for Engineering Design with Negative Data |
| topic | Machine Learning Computational Engineering, Finance, and Science |
| url | https://arxiv.org/abs/2306.15166 |