Enhanced Data-Driven Product Development via Gradient Based Optimization and Conformalized Monte Carlo Dropout Uncertainty Estimation

TL;DR

This paper introduces a novel framework for data-driven product development that combines gradient-based optimization, joint neural networks for multi-property tasks, and a new uncertainty estimation method called Conformalised Monte Carlo Dropout, providing reliable prediction intervals.

Contribution

It presents a new uncertainty estimation technique that offers finite-sample guarantees and integrates it with gradient-based optimization for multi-property product design.

Findings

Matches state-of-the-art performance on real datasets

Provides adaptive, non-uniform prediction intervals

Eliminates retraining when coverage levels change

Abstract

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.