Tentative Program

Tuesday January 8, 2019

Location: Auditorium 6, MH-Building (MH-bygget), University of Tromsø - The Arctic University of Norway


17:00 - 18:00 Pre-workshop Deep Learning tutorial, Michael Kampffmeyer



Wednesday January 9, 2019


Location: UB 132, University Library (Universitetsbibliotek), University of Tromsø - The Arctic University of Norway


08:30 Registration opens


09:00 Opening Robert Jenssen


09:15 Invited Talk: Wojciech Samek


10:00 Coffee break


10:30 Talk 1: Conditional Alignment for Image-to-Image Translation, Federico Baldassarre, KTH - Royal Institute of Technology


10:50 Talk 2: Towards Noise-Resilient Evolution-Strategies, Oswin Krause, University of Copenhagen


11:10 Talk 3: Multilingual Sequence Labeling With One Model, Alan Akbik, Zalando Research


Location: Auditorium 6, MH-Building (MH-bygget), University of Tromsø - The Arctic University of Norway


11:30 Lunch


12:30 Invited Talk: Davide Roverso


13:15 Talk 4: An Exploratory Analysis of Approximated Uncertainty in Multi-Class Image Classification, Sean M Murray, Norsk Regnesentral


13:35 Talk 5: End-to-end Learning for Autonomous Navigation for Agricultural Robots, Marianne Bakken, SINTEF


13:55 Talk 6: Deep Convolutional Networks for Steering an Off-Road Unmanned Ground Vehicle, Johann Dirdal, Norwegian University of Science and Technology


14:15 Coffee break


14:45 Talk 7: An In-depth Study of Classification on Colonoscopy Images, Huamin Ren, Inmeta Consulting AS


15:05 Talk 8: Dense dilated convolutions merging network for semantic mapping of remote sensing images, Qinghui Liu, Norwegian Computing Center


15:25 Northern Lights Talk, Magnar Gullikstad Johnsen


15:50 Poster session (until 16:30)

  • Using a CNN trained on synthetic data for fish species identification, Vaneeda Allken, Institute of Marine Research
  • In-silico Evaluation of Type-1 Diabetes Closed-Loop Control using Deep Reinforcement Learning, Jonas Myhre, UiT The Arctic University of Norway
  • Rethinking Knowledge Graph Propagation for Zero-Shot Learning, Michael Kampffmeyer, UiT The Arctic University of Norway
  • LS-Net: Fast Single Shot Power Line Segment Detector, Van Nhan Nguyen, eSmart Systems
  • Automatic model repairing using deep reinforcement learning, Angela Barriga, Western Norway University of Applied Sciences
  • Avalanche segmentation in SAR data with convolutional neural networks, Filippo Maria Bianchi
  • Machine Learning Derived Input Function in Dynamic 18F-FDG PET, Samuel Kuttner, UiT The Arctic University of Norway
  • An Adversarial Autoencoder Network For Heterogeneous Change Detection, Luigi Tommaso Luppino, UiT The Arctic University of Norway
  • Machine Learning Assisted Quantification of Graphitic Surfaces Exposure to Defined Environments, Chia-Yun Lai, UiT The Arctic University of Norway
  • Uncertainty and Interpretability in Convolutional Neural Networks for Semantic Segmentation of Colorectal Polyps, Kristoffer Wickstrøm, UiT The Arctic University of Norway
  • Recurrent Deep Divergence-based Clustering for simultaneous feature learning and clustering of variable length time series, Daniel Trosten, UiT The Arctic University of Norway
  • Time Series Forecasting with Recurrent Neural Networks in Presence of Missing Data, Changkyu Choi, UiT The Arctic University of Norway
  • (Deep) Generative Models, Rogelio Andrade Mancisidor, Santander Bank
  • Learning Moisture Detection and Fault Detection in Industrial Microwave Drying Process, Yuchong Zhang, Chalmers University of Technology


18:25 Northern light activity (See info below)



Thursday January 10, 2019

Location: UB 132, University Library (Universitetsbibliotek), University of Tromsø - The Arctic University of Norway


09:00 Invited Talk: Marleen de Bruijne


09:45 Coffee break


10:15 Talk 8: Toward Deep Clustering, Ahcène Boubekki, Leuphana University


10:35 Talk 9: Bin Picking of Reflective Steel Parts using a Dual-Resolution Convolutional Neural Network Trained in a Simulated Environment, Marianne Bakken, SINTEF


10:55 Talk 10: Applying deep learning to non-standard image data: a case study on marine acoustics, Olav Brautaset, Norwegian Computing Center


11:15 Talk 11: Contextual Deep Learning Framework for Joint Object Detection & Scene Classification of Ground Level Photos, Seema Chouhan, Oak Ridge National Laboratory


Location: Auditorium 6, MH-Building (MH-bygget), University of Tromsø - The Arctic University of Norway


11:35 Lunch


12:35 Invited Talk: Maurizio Filippone [Slides]


13:20 Talk 12: Recursive Network and Multi-Agent Reinforcement Learning for Smart Houses and Smart Energy, Bernt Bremdal, UiT The Arctic University of Norway


13:40 Talk 13: Detection of electric vehicles in hourly power consumption data, Karoline Ingebrigtsen, SINTEF Energi


14:00 Talk 14: A Deep Learning Approach Towards Prediction of Faults in Wind Turbines, Joyjit Chatterjee, University of Hull


14:20 Coffee break


15:00 Invited Talk: Gitta Kutyniok [Slides]


15:45 Talk 15: Change Detection in Graph Streams by Learning Graph Embeddings on Constant-Curvature Manifolds, Lorenzo Livi,

University of Manitoba


16:05 Closing


16:30 Leaving for dinner


17:00 Workshop dinner (until around 19:30)




Northern Lights Social Event on the 9th of January


We invite you to our Aurora camp where our professional guides will host you for the evening. We will have a fantastic vantage point for the  Aurora Borealis at our centre, only 25 minutes outside of Tromsø. You  will also be introduced to our 300 Huskies and, of course, our puppies.

Relax outside by the camp fire, and keep warm in the Arctic as we wait for the Northern Lights to dance across the sky.

During the visit we provide a warm meal of Finnebiff, a traditional Sami dish, made from Reindeer.


Duration: 4 hours

Pickup/Drop off:

Scandic Ishavshotell 18:30/23:00

Included: Guiding, transport (25 minute each way), dinner, hot drinks, warm clothes