GenPFN: An In-Context Tabular Foundation Model Approach for Augmenting Small Datasets in Engineering
中文摘要
GenPFN利用预训练TabPFN模型,通过上下文学习在工程设计中高效扩充小规模数据集,无需额外训练即可优化并生成高质量设计方案。
English Summary
GenPFN leverages pretrained TabPFN models for in-context learning to augment small engineering datasets, enabling training-free generation and optimization of high-quality design candidates.
arXiv:2611.14804v1 Announce Type: new Abstract: Costly simulations and experiments limit the data available for engineering design. This paper proposes GenPFN, a training-free candidate generator guided by a pretrained tabular Prior-Data Fitted Network (TabPFN) classifier with fixed weights. The context contains evaluated designs labeled desired (Class 1) or undesired (Class 0). GenPFN interpolates between desired designs and refines candidates using gated classifier gradients and Gaussian noise. Each candidate is evaluated using the problem's constraints and objective functions. Evaluated desired designs and their objective values augment the context for regression training. Evaluation spans fourteen selected problems, including a two-parameter benchmark and a 44-parameter hull problem. In paired comparisons, GenPFN with TabPFN v1 achieves higher class satisfaction and lower realism distance than the tested adversarial, variational-autoencoder and diffusion generators. The variational-autoencoder generator provides greater coverage and diversity. Relative to the synthetic minority over-sampling technique (SMOTE), GenPFN provides lower coverage distance and greater diversity. Among…