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

2026-01-02

arXiv.org

AI Summary

This work proposes a novel approach that integrates a joint neural network with Conformalised Monte Carlo Dropout to address the challenges of multi-objective performance optimization and uncertainty quantification in data-driven product development. The method leverages a joint network to model correlations between design parameters and multiple performance objectives, enabling efficient optimization via projected gradient descent. Simultaneously, it combines nested conformal prediction with Monte Carlo Dropout to deliver finite-sample, model-agnostic coverage guarantees and adaptive, non-uniform prediction intervals without requiring model retraining. Experimental results across five real-world datasets demonstrate that the proposed method achieves state-of-the-art performance in multi-objective optimization while offering reliable and well-calibrated uncertainty estimates that can be flexibly adjusted through the desired coverage level.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Stochastic Optimization
• Constraint Satisfaction and Optimization: Constraint Optimization

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

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.

Problem

• Data-Driven Product Development
• Multi-objective Optimization
• Uncertainty Estimation
• Conformal Prediction
• Monte Carlo Dropout

Innovation

• Conformalised Monte Carlo Dropout
• Projected Gradient Descent
• Joint Neural Networks
• Data-Driven Product Development
• Uncertainty Estimation