Learning spectrograms with convolutional spectral kernels

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

49 Downloads (Pure)


We introduce the convolutional spectral kernel (CSK), a novel family of non-stationary, nonparametric covariance kernels for Gaussian process (GP) models, derived from the convolution between two imaginary radial basis functions. We present a principled framework to interpret CSK, as well as other deep probabilistic models, using approximated Fourier transform, yielding a concise representation of input-frequency spectrogram. Observing through the lens of the spectrogram, we provide insight on the interpretability of deep models. We then infer the functional hyperparameters using scalable variational and MCMC methods. On small- and medium-sized spatiotemporal datasets, we demonstrate improved generalization of GP models when equipped with CSK, and their capability to extract non-stationary periodic patterns.
Original languageEnglish
Title of host publicationThe 23rd International Conference on Artificial Intelligence and Statistics
EditorsS Chiappa, R Calandra
Number of pages10
Publication statusPublished - 2020
MoE publication typeA4 Conference publication
EventInternational Conference on Artificial Intelligence and Statistics - Palermo, Italy
Duration: 3 Jun 20205 Jun 2020
Conference number: 23

Publication series

NameProceedings of Machine Learning Research
ISSN (Print)2640-3498


ConferenceInternational Conference on Artificial Intelligence and Statistics
Abbreviated titleAISTATS


Dive into the research topics of 'Learning spectrograms with convolutional spectral kernels'. Together they form a unique fingerprint.

Cite this