Transfer learning based few-shot classification using optimal transport mapping from preprocessed latent space of backbone neural network

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F21%3A00356449" target="_blank" >RIV/68407700:21240/21:00356449 - isvavai.cz</a>

Result on the web

<a href="https://proceedings.mlr.press/v140/chobola21a.html" target="_blank" >https://proceedings.mlr.press/v140/chobola21a.html</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Transfer learning based few-shot classification using optimal transport mapping from preprocessed latent space of backbone neural network

Original language description

The MetaDL Challenge 2020 focused on image classification tasks in few-shot settings. This paper describes second best submission in the competition. Our meta learning approach modifies the distribution of classes in a latent space produced by a backbone network for each class in order to better follow the Gaussian distribution. After this operation which we call Latent Space Transform algorithm, centers of classes are further aligned in an iterative fashion of the Expectation Maximisation algorithm to utilize information in unlabeled data that are often provided on top of few labelled instances. For this task, we utilize optimal transport mapping using the Sinkhorn algorithm. Our experiments show that this approach outperforms previous works as well as other variants of the algorithm, using K-Nearest Neighbour algorithm, Gaussian Mixture Models, etc.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

<a href="/en/project/GA18-18080S" target="_blank" >GA18-18080S: Fusion-Based Knowledge Discovery in Human Activity Data</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2021

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

AAAI Workshop on Meta-Learning and MetaDL Challenge

ISBN

—

ISSN

2640-3498

e-ISSN

2640-3498

Number of pages

9

Pages from-to

29-37

Publisher name

Proceedings of Machine Learning Research

Place of publication

—

Event location

Virtuální

Event date

Feb 8, 2021

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—