Wasserstein dropout

Autor: Joachim Sicking, Maram Akila, Maximilian Pintz, Tim Wirtz, Stefan Wrobel, Asja Fischer

Publikováno v: Machine Learning.

Despite of its importance for safe machine learning, uncertainty quantification for neural networks is far from being solved. State-of-the-art approaches to estimate neural uncertainties are often hybrid, combining parametric models with explicit or

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c245f5e6777583b4a742dfb327060892
https://doi.org/10.1007/s10994-022-06230-8

Bringing Light Into the Dark: A Large-Scale Evaluation of Knowledge Graph Embedding Models Under a Unified Framework

Autor: Asja Fischer, Charles Tapley Hoyt, Jens Lehmann, Volker Tresp, Mikhail Galkin, Max Berrendorf, Laurent Vermue, Mehdi Ali, Sahand Sharifzadeh

Publikováno v: Ali, M, Berrendorf, M, Hoyt, C T, Vermue, L, Galkin, M, Sharifzadeh, S, Fischer, A, Tresp, V & Lehmann, J 2022, ' Bringing Light Into the Dark: A Large-scale Evaluation of Knowledge Graph Embedding Models under a Unified Framework ', Ieee Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 12, pp. 8825-8845 . https://doi.org/10.1109/TPAMI.2021.3124805

The heterogeneity in recently published knowledge graph embedding models' implementations, training, and evaluation has made fair and thorough comparisons difficult. In order to assess the reproducibility of previously published results, we re-implem

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::450a8ab4d4fb1326f2b5b2017153e6fd
https://pubmed.ncbi.nlm.nih.gov/34735335

SmoothLRP: Smoothing LRP by Averaging over Stochastic Input Variations

Autor: Ben Luis Hack, Axel Mosig, Sina Däubener, Asja Fischer, Arne P. Raulf

Publikováno v: ESANN 2021 proceedings.

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::7fb92494c817d68be3e0ff17c7db1d30
https://doi.org/10.14428/esann/2021.es2021-139

HORUS-NER: A Multimodal Named Entity Recognition Framework for Noisy Data

Autor: Diego Esteves, Asja Fischer, Jens Lehmann, José Marcelino, Piyush Chawla

Publikováno v: Advances in Intelligent Data Analysis XIX ISBN: 9783030742508
IDA

Recent work based on Deep Learning presents state-of-the-art (SOTA) performance in the named entity recognition (NER) task. However, such models still have the performance drastically reduced in noisy data (e.g., social media, search engines), when c

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::e423091ee259b40a104684c237e98cc1
https://doi.org/10.1007/978-3-030-74251-5_8

Approaches to Uncertainty Quantification in Federated Deep Learning

Autor: Florian Linsner, Linara Adilova, Sina Däubener, Michael Kamp, Asja Fischer

Publikováno v: Communications in Computer and Information Science ISBN: 9783030937355

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::30c7f4cdfb1e5147e747cd798f18c433
https://doi.org/10.1007/978-3-030-93736-2_12