Neural Turing Machine for Sequential Learning of Human Mobility Patterns

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F16%3A00302589" target="_blank" >RIV/68407700:21240/16:00302589 - isvavai.cz</a>

Výsledek na webu

<a href="http://ieeexplore.ieee.org/document/7727551/" target="_blank" >http://ieeexplore.ieee.org/document/7727551/</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/IJCNN.2016.7727551" target="_blank" >10.1109/IJCNN.2016.7727551</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Neural Turing Machine for Sequential Learning of Human Mobility Patterns

Popis výsledku v původním jazyce

The capacity of recurrent neural networks to learn complex sequential patterns is improving. Recent developments such as Clockwork RNN, Stack RNN, Memory networks and Neural Turing Machine all aim to increase long-term memory capacity of recurrent neural networks. In this study, we investigate properties of Neural Turing Machine, compare it with ensembles of Stack RNN on artificial benchmarks and applied it to learn human mobility patterns. We show, that Neural Turing Machine based predictor outperformed not only n-gram based prediction, but also neighborhood based predictor, that was designed to solve this particular problem. Our models will be deployed in anti-drug police department to predict mobility of suspects.

Název v anglickém jazyce

Neural Turing Machine for Sequential Learning of Human Mobility Patterns

Popis výsledku anglicky

The capacity of recurrent neural networks to learn complex sequential patterns is improving. Recent developments such as Clockwork RNN, Stack RNN, Memory networks and Neural Turing Machine all aim to increase long-term memory capacity of recurrent neural networks. In this study, we investigate properties of Neural Turing Machine, compare it with ensembles of Stack RNN on artificial benchmarks and applied it to learn human mobility patterns. We show, that Neural Turing Machine based predictor outperformed not only n-gram based prediction, but also neighborhood based predictor, that was designed to solve this particular problem. Our models will be deployed in anti-drug police department to predict mobility of suspects.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2016

Kód důvěrnosti údajů

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

Název statě ve sborníku

2016 International Joint Conference on Neural Networks (IJCNN)

ISBN

978-1-5090-0620-5

ISSN

2161-4407

e-ISSN

—

Počet stran výsledku

8

Strana od-do

2790-2797

Název nakladatele

American Institute of Physics and Magnetic Society of the IEEE

Místo vydání

San Francisco

Místo konání akce

Vancouver

Datum konání akce

24. 7. 2016

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—