Back to results

University of Cambridge

Random Features for Efficient Attention Approximation

Abstract

dc:description.abstract

Transformers are, perhaps, the most widespread architectures in today's landscape of deep learning. They, however, do not scale well with long sequences, resulting in O(L^2) computational complexity for the sequence length L. In this thesis, we propose a holistic approach for reducing this complexity to linear O(L) via an unbiased approximation of the softmax kernel appearing in self-attention, the main component in the Transformer backbone. The obtained efficient Transformer architecture is referred to as Performer. Compared to other developments in the area of efficient Transformers, Performer is theory-grounded and the problem of long-sequence processing can be reduced to a theoretical derivation of random features with certain properties minimizing the variance of the approximation. This thesis describes an evolution of mechanisms for random-feature approximation: from the so-called FAVOR, to FAVOR+ and, finally, to FAVOR++. The FAVOR++ mechanism has the tightest concentration properties and the best performance in practice. On the way to FAVOR++, we also discuss several extensions of the proposed efficient self-attention mechanism, among which are masked self-attention, sub-linear memory Performers, generalized attention, and more. For each proposed method, this thesis contains empirical evaluations in real-life large-scale learning setups and thorough theoretical analyses with proofs.

Degree

thesis:*
Name dc:type.qualificationname
Doctor of Philosophy (PhD)
Level dc:type.qualificationlevel
Doctoral
Grantor dc:publisher.institution
University of Cambridge
Year dc:date.issued
2022

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Likhosherstov, Valerii
Advisor dc:contributor.advisor
  • Weller, Adrian

Subjects

dc:subject × 3

Rights

dc:rights
Language dc:language
eng

Identifiers

dc:identifier.*
DOI dc:identifier.doi
https://doi.org/10.17863/CAM.96501
OAI identifier oai:identifier
oai:www.repository.cam.ac.uk:1810/349393

Chain of custody

source
Harvested from
Cambridge University
Base URL
api.repository.cam.ac.uk/server/oai/request
Last updated
2026-07-22
Source record
OAI-PMH GetRecord
citation

Likhosherstov, Valerii. Random Features for Efficient Attention Approximation. Doctoral thesis, University of Cambridge, 2022. https://doi.org/10.17863/CAM.96501