Leveraging Active Subspaces to Capture Epistemic Model Uncertainty in Deep Generative Models for Molecular Design

Abstract

Deep generative models have been accelerating the inverse design process in material and drug design. Unlike their counterpart property predictors in typical molecular design frameworks, generative molecular design models have seen fewer efforts on uncertainty quantification (UQ) due to computational challenges in Bayesian inference posed by their large number of parameters. In this work, we focus on the junction-tree variational autoencoder (JT-VAE), a popular model for generative molecular design, and address this issue by leveraging the low dimensional active subspace to capture the uncertainty in the model parameters. Specifically, we approximate the posterior distribution over the active subspace parameters to estimate the epistemic model uncertainty in an extremely high dimensional parameter space. The proposed UQ scheme does not require alteration of the model architecture, making it readily applicable to any pre-trained model. Our experiments demonstrate the efficacy of the AS-based UQ and its potential impact on molecular optimization by exploring the model diversity under epistemic uncertainty.

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• Variational Inference: A Review for Statisticians
• Automatic chemical design using a data-driven continuous representation of molecules
• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
• Weight Uncertainty in Neural Networks
• Junction Tree Variational Autoencoder for Molecular Graph Generation
• Optimization of Molecules via Deep Reinforcement Learning
• Bayesian Compression for Deep Learning
• Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development
• What Are Bayesian Neural Network Posteriors Really Like?
• Multi-Objective Molecule Generation using Interpretable Substructures
• Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization
• Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retraining
• Uncertainty Quantification for Molecular Property Predictions with Graph Neural Architecture Search
• Rethinking Parameter Counting in Deep Models: Effective Dimensionality Revisited
• Improving black-box optimization in VAE latent space using decoder uncertainty
• Layer Adaptive Node Selection in Bayesian Neural Networks: Statistical Guarantees and Implementation Details