User: Guest  Login
If you experience problems opening the document, please try this link.
Original title:
Modeling Continuous-time Event Data with Neural Temporal Point Processes
Translated title:
Modellierung von zeitkontinuierlichen Ereignisdaten mit neuronalen temporalen Punktprozessen
Author:
Shchur, Oleksandr
Year:
2022
Document type:
Dissertation
Faculty/School:
TUM School of Computation, Information and Technology
Advisor:
Günnemann, Stephan (Prof. Dr.)
Referee:
Günnemann, Stephan (Prof. Dr.); Linderman, Scott (Prof. Dr.)
Language:
en
Subject group:
DAT Datenverarbeitung, Informatik
Keywords:
machine learning, probabilistic modeling, event data, temporal point process
TUM classification:
DAT 600; DAT 703; DAT 708
Abstract:
Temporal point processes (TPPs) provide a natural framework for modeling continuous-time event data such as earthquake catalogs in seismology or spike trains in neuroscience. Unlike conventional TPP models, neural TPPs are able to capture complex patterns present in real-world event data. The two main themes of this thesis are design of flexible, tractable and efficient neural TPP models, and their applications to real-world problems.
Translated abstract:
Temporale Punktprozesse (TPPs) bilden den natürlichen Rahmen zur Modellierung von Ereignisfolgen wie Erdbebenkataloge in der Seismologie oder Spike Trains in der Neurowissenschaft. Im Gegensatz zu konventionellen TPP-Modelle können neuronale TPPs komplexe Muster aus realen Ereignisdaten erfassen. Den Schwerpunkt dieser Arbeit bilden der Entwurf flexibler, interpretierbarer und effizienter neuronaler TPP-Modelle und die Anwendung dieser auf reale Probleme.
WWW:
https://mediatum.ub.tum.de/?id=1662914
Date of submission:
04.07.2022
Oral examination:
12.12.2022
File size:
4422511 bytes
Pages:
159
Urn (citeable URL):
https://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:91-diss-20221212-1662914-1-9
Last change:
20.01.2023
 BibTeX