Adaptive Deep Kernel Learning
This work addresses few-shot learning for complex task distributions, specifically in drug discovery, and is incremental as it extends deep kernel learning to multi-kernel settings.
The paper tackled the problem of few-shot regression by learning a family of kernels instead of a single kernel, enabling task-specific kernel selection during inference. The result was demonstrated through comparisons with state-of-the-art algorithms on real-world drug discovery tasks.
Deep kernel learning provides an elegant and principled framework for combining the structural properties of deep learning algorithms with the flexibility of kernel methods. By means of a deep neural network, we learn a parametrized kernel operator that can be combined with a differentiable kernel algorithm during inference. While previous work within this framework has focused on learning a single kernel for large datasets, we learn a kernel family for a variety of few-shot regression tasks. Compared to single deep kernel learning, our algorithm enables the identification of the appropriate kernel for each task during inference. As such, it is well adapted for complex task distributions in a few-shot learning setting, which we demonstrate by comparing against existing state-of-the-art algorithms using real-world, few-shot regression tasks related to the field of drug discovery.