Modeling Structure with Undirected Neural Networks

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F22%3AJDCSP774" target="_blank" >RIV/00216208:11320/22:JDCSP774 - isvavai.cz</a>

Result on the web

<a href="https://proceedings.mlr.press/v162/mihaylova22a.html" target="_blank" >https://proceedings.mlr.press/v162/mihaylova22a.html</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Modeling Structure with Undirected Neural Networks

Original language description

Neural networks are powerful function estimators, leading to their status as a paradigm of choice for modeling structured data. However, unlike other structured representations that emphasize the modularity of the problem {–} e.g., factor graphs {–} neural networks are usually monolithic mappings from inputs to outputs, with a fixed computation order. This limitation prevents them from capturing different directions of computation and interaction between the modeled variables. In this paper, we combine the representational strengths of factor graphs and of neural networks, proposing undirected neural networks (UNNs): a flexible framework for specifying computations that can be performed in any order. For particular choices, our proposed models subsume and extend many existing architectures: feed-forward, recurrent, self-attention networks, auto-encoders, and networks with implicit layers. We demonstrate the effectiveness of undirected neural architectures, both unstructured and structured, on a range of tasks: tree-constrained dependency parsing, convolutional image classification, and sequence completion with attention. By varying the computation order, we show how a single UNN can be used both as a classifier and a prototype generator, and how it can fill in missing parts of an input sequence, making them a promising field for further research.

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

—

Continuities

—

Publication year

2022

Confidentiality

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

Article name in the collection

Proceedings of the 39th International Conference on Machine Learning

ISBN

—

ISSN

2640-3498

e-ISSN

—

Number of pages

18

Pages from-to

468-485

Publisher name

PMLR

Place of publication

—

Event location

Baltimore, Maryland, USA

Event date

Jan 1, 2022

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—