Meta-learning Adaptive Deep Kernel Gaussian Processes for Molecular Property Prediction
This work addresses drug discovery challenges by improving predictive accuracy for molecular properties, though it appears incremental as it builds on existing deep kernel learning methods.
The paper tackles the problem of few-shot and out-of-domain molecular property prediction by proposing ADKF-IFT, a meta-learning framework for deep kernel Gaussian processes, which significantly outperforms previous state-of-the-art methods on real-world tasks.
We propose Adaptive Deep Kernel Fitting with Implicit Function Theorem (ADKF-IFT), a novel framework for learning deep kernel Gaussian processes (GPs) by interpolating between meta-learning and conventional deep kernel learning. Our approach employs a bilevel optimization objective where we meta-learn generally useful feature representations across tasks, in the sense that task-specific GP models estimated on top of such features achieve the lowest possible predictive loss on average. We solve the resulting nested optimization problem using the implicit function theorem (IFT). We show that our ADKF-IFT framework contains previously proposed Deep Kernel Learning (DKL) and Deep Kernel Transfer (DKT) as special cases. Although ADKF-IFT is a completely general method, we argue that it is especially well-suited for drug discovery problems and demonstrate that it significantly outperforms previous state-of-the-art methods on a variety of real-world few-shot molecular property prediction tasks and out-of-domain molecular property prediction and optimization tasks.