Login / Signup
TaylorShift: Shifting the Complexity of Self-Attention from Squared to Linear (and Back) using Taylor-Softmax.
Tobias Christian Nauen
Sebastian Palacio
Andreas Dengel
Published in:
CoRR (2024)
Keyphrases
</>
linear complexity
least squares
linear model
visual attention
computer vision
case study
focus of attention
error tolerance
data sets
activation function
space complexity
closed form
worst case
computational cost
mobile robot
evolutionary algorithm
computational complexity
multiscale
genetic algorithm