Scalable Gaussian Processes on Discrete Domains
This work addresses scalability issues in probabilistic kernel methods for domains like biological sequences, offering incremental improvements in inducing point selection.
The paper tackled the challenge of selecting inducing points for scalable Gaussian Processes on discrete domains, finding that simulated annealing performs competitively with support vector machines and full Gaussian processes, achieving comparable predictive performance on synthetic and real-world DNA sequence data.
Kernel methods on discrete domains have shown great promise for many challenging data types, for instance, biological sequence data and molecular structure data. Scalable kernel methods like Support Vector Machines may offer good predictive performances but do not intrinsically provide uncertainty estimates. In contrast, probabilistic kernel methods like Gaussian Processes offer uncertainty estimates in addition to good predictive performance but fall short in terms of scalability. While the scalability of Gaussian processes can be improved using sparse inducing point approximations, the selection of these inducing points remains challenging. We explore different techniques for selecting inducing points on discrete domains, including greedy selection, determinantal point processes, and simulated annealing. We find that simulated annealing, which can select inducing points that are not in the training set, can perform competitively with support vector machines and full Gaussian processes on synthetic data, as well as on challenging real-world DNA sequence data.