用梯度优化与不确定性估计，加速多属性产品设计迭代
Enhanced Data-Driven Product Development via Gradient Based Optimization and Conformalized Monte Carlo Dropout Uncertainty Estimation
- 结合梯度下降与联合神经网络，同步优化多个相关产品属性
- 在5个真实数据集上达到顶尖性能，预测区间可自适应调整
- 无需重训即可改变置信水平，适合需要可靠决策的工业研发
数据驱动的产品开发（DDPD）利用数据学习产品设计规格与性能之间的关系。为发现更优设计，我们在过往实验数据上训练神经网络，并采用投影梯度下降法寻找能最大化性能的输入特征。由于许多产品需同时优化多个相关属性，本框架采用联合神经网络捕捉目标间的依赖关系。此外，我们引入基于蒙特卡洛丢弃的共形化方法（ConfMC），该方法将嵌套共形预测与蒙特卡洛丢弃结合，在数据可交换性假设下提供模型无关、有限样本覆盖保证。在五个真实世界数据集上的大量实验表明，该方法性能媲美现有最优方案，同时具备自适应、非均匀的预测区间，且在调整置信水平时无需重新训练。
原文摘要 · Abstract (English)
Data-Driven Product Development (DDPD) leverages data to learn the relationship between product design specifications and resulting properties. To discover improved designs, we train a neural network on past experiments and apply Projected Gradient Descent to identify optimal input features that maximize performance. Since many products require simultaneous optimization of multiple correlated properties, our framework employs joint neural networks to capture interdependencies among targets. Furthermore, we integrate uncertainty estimation via \emph{Conformalised Monte Carlo Dropout} (ConfMC), a novel method combining Nested Conformal Prediction with Monte Carlo dropout to provide model-agnostic, finite-sample coverage guarantees under data exchangeability. Extensive experiments on five real-world datasets show that our method matches state-of-the-art performance while offering adaptive, non-uniform prediction intervals and eliminating the need for retraining when adjusting coverage levels.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。