About

I am currently a Research Assistant at the Technical University of Denmark where I lead the student-based research organization, Deep Learning Copenhagen, together with Jose J.A. Armenteros (Ph.D. student, Bioinformatics). I previously did a one-year internship at Salesforce Research supervised by Richard Socher. I got my M.Sc. honors from the Technical University of Denmark supervised by Ole Winther. I am interested in Natural Language Processing, Generalization, Reinforcement Learning, BioInformatics, and ChemInformatics.

alexander@herhjemme.dk Google Scholar Profile Resume GitHub LinkedIn Twitter

Publications

Neural Arithmetic Units
A. Madsen, A.R. Johansen
ICLR (Spotlight), 2020
PDF Abstract Bibtex Website

Language modelling for biological sequences curated datasets and baselines
J.J.A. Armenteros*, A.R. Johansen*, O. Winther, H. Nielsen (*equal contribution)
Preprint and website, 2019
PDF Abstract Bibtex Website

Autoencoding undirected molecular graphs with neural networks
J.J.W. Olsen, P.E. Christensen, M.H. Hansen, A.R. Johansen
Under review Journal of Chemical Information and Modeling, 2019
PDF Abstract Bibtex

Prediction of GPI-Anchored proteins with pointer neural networks
M.H. Gislason, H. Nielsen, J.J.A. Armenteros*, A.R. Johansen* (*equal contribution)
Under review PROTEINS Structure, Function, and Bioinformatics, 2019
PDF Abstract Bibtex Website

Measuring Arithmetic Extrapolation Performance
A. Madsen, A.R. Johansen
SEDL Workshop @ NeurIPS, 2019
PDF Abstract Bibtex Code Website

An introduction to deep learning on biological sequence data - examples and solutions
V.I. Jurtz, A.R. Johansen, M. Nielsen, J.J.A. Armenteros, H. Nielsen, C.K. Sønderby, O. Winther, S.K. Sønderby
Oxford Bioinformatics, 2017
Abstract Bibtex Code

Deep Recurrent Conditional Random Field Network for Protein Secondary Prediction
A.R. Johansen, C.K. Sønderby, S.K. Sønderby, O. Winther
ACM BCB, 2017
PDF Abstract Bibtex Code

A deep learning approach to adherence detection for type 2 diabetics
A. Mohebbi, T.B. Aradóttir, A.R. Johansen, H. Bengtsson, M. Fraccaro, and M. Mørup
IEEE EMBC, 2017
Abstract Bibtex

Learning when to skim and when to read
A.R. Johansen, R. Socher
REPL4NLP Workshop @ ACL, 2017
PDF Abstract Bibtex

Neural Machine Translation with Characters and Hierarchical Encoding
A.R. Johansen, J.M. Hansen, E.K. Obeid, C.K. Sønderby, O. Winther
RNN Symposium @ NIPS, 2016
PDF Abstract Bibtex Code

Epileptiform spike detection via convolutional neural networks
A.R. Johansen, J. Jin, T. Maszczyk, J. Dauwels, S.S. Cash, M.B. Westover
IEEE ICASSP, 2016
Abstract Bibtex Code


Teaching

Below you can find a list over courses I have taught in.

DTU course 02456 Deep learning
Head teaching assistant, designed programming exercises in PyTorch
Fall, 2019
Code Website

DTU special course Deep reinforcement learning
Co-intructor, designed and evaluated course
June, 2019

DTU special course Introduction to reinforcement learning
Intructor, designed and evaluated course
Spring, 2019

DTU course 02456 Deep learning
Head teaching assistant, developed programming exercises in PyTorch
Fall, 2018
Code Website

DTU course 02456 Deep learning
Teaching assistant, developed programming exercises in TensorFlow
Fall, 2016
Code Website

Nvidia Deep Learning using TensorFlow
Exercise instructor, developed programming exercises in TensorFlow
September, 2016
Code


Supervising

As a part of my student lab with Jose J.A. Armenteros (DTU Bioinformatics) we supervise graduate students at the Technical University of Denmark. We supervise special courses (5-10 ECTS) and Master Thesis' (+30 ECTS). If you are interested in joining our lab please write me an email with you resume and detail your background in machine learning. We meet every Tuesday at 09:00 for football and 10:30-12:00 for the weekly research meeting. Below are listed titles of the M.Sc. Thesis' (+30 ECTS) I have co-supervised.

Multi-label prediction of protein subcellular localization using deep learning
Morten Skovsted
Autumn, 2019

Generalized autoregressive pretraining for improved understanding of proteins
Magnus Nolsøe and Frederik Wollesen Andersen
Autumn, 2019

Partially autoregressive language modelling and editing of discrete sequences' thesis
Daniel Horvath
Autumn, 2019

Prediction of sorting signals in eukaryotic proteins using deep learning
Magnús Halldór Gíslason
Autumn, 2019

Visual question answering using structured exploration and reinforcement learning
Jacob Johansen
Autumn, 2019

Evaluation of contextual molecular representations on predicting structural properties in binarized molecules
Jeppe Waarkjær Olsen
Autumn, 2019

Evaluation of contextual amino acid representations on the prediction of N-terminal targeting peptides based on deep learning
Silas Pihl
Autumn, 2019

Formal Language in Natural Language Processing
Raja Shan Zaker Krenn
Spring, 2019

Predicting Recombinant Gene Expression in Bacillus using Deep Learning Techniques
Hannah-Marie Martiny
Spring, 2019

Evaluation of Pre-trained Amino Acid Embeddings in Protein Prediction Tasks
Mikkel Møller Brusen and Gustav Madslund
Spring, 2019
Code


Community

To build a strong deep learning community in Denmark I organize events for students with machine learning researchers and interested companies. So far this has led to 7 events with +1.5k student participants and sponsors such as Nordea, KPMG, Novozymes, Oticon, and Novo Nordisk.

Deep Learning Copenhagen
Organizing events centered around maching learning technologies
2018-present
Meetup