Variational Auto-Regressive Gaussian Processes for Continual Learning

Sanyam Kapoor, Theofanis Karaletsos, Thang D Bui
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:5290-5300, 2021.

Abstract

Through sequential construction of posteriors on observing data online, Bayes’ theorem provides a natural framework for continual learning. We develop Variational Auto-Regressive Gaussian Processes (VAR-GPs), a principled posterior updating mechanism to solve sequential tasks in continual learning. By relying on sparse inducing point approximations for scalable posteriors, we propose a novel auto-regressive variational distribution which reveals two fruitful connections to existing results in Bayesian inference, expectation propagation and orthogonal inducing points. Mean predictive entropy estimates show VAR-GPs prevent catastrophic forgetting, which is empirically supported by strong performance on modern continual learning benchmarks against competitive baselines. A thorough ablation study demonstrates the efficacy of our modeling choices.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-kapoor21b, title = {Variational Auto-Regressive Gaussian Processes for Continual Learning}, author = {Kapoor, Sanyam and Karaletsos, Theofanis and Bui, Thang D}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {5290--5300}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/kapoor21b/kapoor21b.pdf}, url = {https://proceedings.mlr.press/v139/kapoor21b.html}, abstract = {Through sequential construction of posteriors on observing data online, Bayes’ theorem provides a natural framework for continual learning. We develop Variational Auto-Regressive Gaussian Processes (VAR-GPs), a principled posterior updating mechanism to solve sequential tasks in continual learning. By relying on sparse inducing point approximations for scalable posteriors, we propose a novel auto-regressive variational distribution which reveals two fruitful connections to existing results in Bayesian inference, expectation propagation and orthogonal inducing points. Mean predictive entropy estimates show VAR-GPs prevent catastrophic forgetting, which is empirically supported by strong performance on modern continual learning benchmarks against competitive baselines. A thorough ablation study demonstrates the efficacy of our modeling choices.} }
Endnote
%0 Conference Paper %T Variational Auto-Regressive Gaussian Processes for Continual Learning %A Sanyam Kapoor %A Theofanis Karaletsos %A Thang D Bui %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-kapoor21b %I PMLR %P 5290--5300 %U https://proceedings.mlr.press/v139/kapoor21b.html %V 139 %X Through sequential construction of posteriors on observing data online, Bayes’ theorem provides a natural framework for continual learning. We develop Variational Auto-Regressive Gaussian Processes (VAR-GPs), a principled posterior updating mechanism to solve sequential tasks in continual learning. By relying on sparse inducing point approximations for scalable posteriors, we propose a novel auto-regressive variational distribution which reveals two fruitful connections to existing results in Bayesian inference, expectation propagation and orthogonal inducing points. Mean predictive entropy estimates show VAR-GPs prevent catastrophic forgetting, which is empirically supported by strong performance on modern continual learning benchmarks against competitive baselines. A thorough ablation study demonstrates the efficacy of our modeling choices.
APA
Kapoor, S., Karaletsos, T. & Bui, T.D.. (2021). Variational Auto-Regressive Gaussian Processes for Continual Learning. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:5290-5300 Available from https://proceedings.mlr.press/v139/kapoor21b.html.

Related Material