Erlend Aune
About
I have a strong interest for creative use of machine learning/statistical modeling for specific applications. Examples of modeling frameworks and applications I work within are:
- Generative modeling (diffusion modeling, masked modeling)
- Representation learning
- Active Learning
- Modeling of time series
- Forecasting
- Synthetic data generation
- Imputation/denoising
- Irregular sampling
- Machine learning with food
- Content extraction and relation extraction from text using machine learning
My primary position is Research Scientist at Abelee, and I am also the founder (with Gudmund Hermansen) of Oslo Data Science Meetup.
Research
Publications
2023
-
Lee, Daesoo;
Malacarne, Sara;
Aune, Erlend.
(2023)
Vector Quantized Time Series Generation with a Bidirectional Prior Model.
Proceedings of Machine Learning Research (PMLR)
Academic article
-
Haugsdal, Espen;
Aune, Erlend;
Ruocco, Massimiliano.
(2023)
Persistence Initialization: a novel adaptation of the Transformer architecture for time series forecasting.
Applied intelligence (Boston)
Academic article
2022
-
Lee, Daesoo;
Aune, Erlend;
Langet, Nadege;
Eidsvik, Jo.
(2022)
Ensemble and Self-supervised Learning for Improved Classification of Seismic Signals from the Åknes Rockslope.
Mathematical Geosciences
Academic literature review
2019
-
Vassøy, Bjørnar;
Ruocco, Massimiliano;
de Souza da Silva, Eliezer;
Aune, Erlend.
(2019)
Time is of the essence: A joint Hierarchical RNN and Point Process model for time and item predictions.
Association for Computing Machinery (ACM)
Academic chapter/article/Conference paper
2013
-
Aune, Erlend;
Eidsvik, Jo;
Pokern, Y.
(2013)
Iterative numerical methods for sampling from high dimensional Gaussian distributions.
Statistics and computing
Academic article
-
Aune, Erlend;
Eidsvik, Jo;
Ursin, Bjørn.
(2013)
Three-dimensional non-stationary and non-linear isotropic AVA inversion.
Geophysical Journal International
Academic article
2012
-
Aune, Erlend.
(2012)
Computation and modeling for high dimensional Gaussian distributions.
Norges teknisk-naturvitenskapelige universitet
Norges teknisk-naturvitenskapelige universitet
Doctoral dissertation
Journal publications
-
Lee, Daesoo;
Malacarne, Sara;
Aune, Erlend.
(2023)
Vector Quantized Time Series Generation with a Bidirectional Prior Model.
Proceedings of Machine Learning Research (PMLR)
Academic article
-
Haugsdal, Espen;
Aune, Erlend;
Ruocco, Massimiliano.
(2023)
Persistence Initialization: a novel adaptation of the Transformer architecture for time series forecasting.
Applied intelligence (Boston)
Academic article
-
Lee, Daesoo;
Aune, Erlend;
Langet, Nadege;
Eidsvik, Jo.
(2022)
Ensemble and Self-supervised Learning for Improved Classification of Seismic Signals from the Åknes Rockslope.
Mathematical Geosciences
Academic literature review
-
Aune, Erlend;
Eidsvik, Jo;
Pokern, Y.
(2013)
Iterative numerical methods for sampling from high dimensional Gaussian distributions.
Statistics and computing
Academic article
-
Aune, Erlend;
Eidsvik, Jo;
Ursin, Bjørn.
(2013)
Three-dimensional non-stationary and non-linear isotropic AVA inversion.
Geophysical Journal International
Academic article
Part of book/report
-
Vassøy, Bjørnar;
Ruocco, Massimiliano;
de Souza da Silva, Eliezer;
Aune, Erlend.
(2019)
Time is of the essence: A joint Hierarchical RNN and Point Process model for time and item predictions.
Association for Computing Machinery (ACM)
Academic chapter/article/Conference paper
Report
-
Aune, Erlend.
(2012)
Computation and modeling for high dimensional Gaussian distributions.
Norges teknisk-naturvitenskapelige universitet
Norges teknisk-naturvitenskapelige universitet
Doctoral dissertation
Media
2021
-
Programme participationSaplacan, Diana; Foldnes, Njål; Aune, Erlend; Dahl, Ida; Voigt, Jakob Michael; Goodwin, Morten. (2021) AI & pedagogics. Webinar - Norwegian Consortium of Artificial Intelligence Webinar - Norwegian Consortium of Artificial Intelligence [Internet] 2021-10-01
2016
-
InterviewAune, Erlend; Stenvik, Lars Fredrik. (2016) Mektig imponert over jernviljen. Innherred Innherred [Newspaper] 2016-10-06