Abstract
This paper studies few-shot learning via representation learning, where one
uses $T$ source tasks with $n_1$ data per task to learn a representation in
order to reduce the sample complexity of a target task for which there is only
$n_2 (n_1)$ data. Specifically, we focus on the setting where there exists
a good common representation between source and target, and our goal is
to understand how much of a sample size reduction is possible. First, we study
the setting where this common representation is low-dimensional and provide a
fast rate of $Ołeft(\mathcalCłeft(\Phi\right)n_1T +
kn_2\right)$; here, $\Phi$ is the representation function class,
$Cłeft(\Phi\right)$ is its complexity measure, and $k$ is the
dimension of the representation. When specialized to linear representation
functions, this rate becomes $Ołeft(dkn_1T + kn_2\right)$
where $d (k)$ is the ambient input dimension, which is a substantial
improvement over the rate without using representation learning, i.e. over the
rate of $Ołeft(dn_2\right)$. Second, we consider the setting where
the common representation may be high-dimensional but is capacity-constrained
(say in norm); here, we again demonstrate the advantage of representation
learning in both high-dimensional linear regression and neural network
learning. Our results demonstrate representation learning can fully utilize all
$n_1T$ samples from source tasks.
Description
[2002.09434] Few-Shot Learning via Learning the Representation, Provably
Links and resources
Tags
community