From block-Toeplitz matrices to differential equations on graphs: towards a general theory for scalable masked Transformers

Krzysztof Choromanski, Han Lin, Haoxian Chen, Tianyi Zhang, Arijit Sehanobish, Valerii Likhosherstov, Jack Parker-Holder, Tamás Sarlós, Adrian Weller, Thomas Weingarten. From block-Toeplitz matrices to differential equations on graphs: towards a general theory for scalable masked Transformers. In Kamalika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepesvári, Gang Niu 0001, Sivan Sabato, editors, International Conference on Machine Learning, ICML 2022, 17-23 July 2022, Baltimore, Maryland, USA. Volume 162 of Proceedings of Machine Learning Research, pages 3962-3983, PMLR, 2022. [doi]

Authors

Krzysztof Choromanski

This author has not been identified. Look up 'Krzysztof Choromanski' in Google

Han Lin

This author has not been identified. Look up 'Han Lin' in Google

Haoxian Chen

This author has not been identified. Look up 'Haoxian Chen' in Google

Tianyi Zhang

This author has not been identified. Look up 'Tianyi Zhang' in Google

Arijit Sehanobish

This author has not been identified. Look up 'Arijit Sehanobish' in Google

Valerii Likhosherstov

This author has not been identified. Look up 'Valerii Likhosherstov' in Google

Jack Parker-Holder

This author has not been identified. Look up 'Jack Parker-Holder' in Google

Tamás Sarlós

This author has not been identified. Look up 'Tamás Sarlós' in Google

Adrian Weller

This author has not been identified. Look up 'Adrian Weller' in Google

Thomas Weingarten

This author has not been identified. Look up 'Thomas Weingarten' in Google