4 papers
Improved optimization strategies for deep Multi-Task Networks
Lucas Pascal, Pietro Michiardi, Xavier Bost +2
In Multi-Task Learning (MTL), it is a common practice to train multi-task networks by optimizing an objective function, which is a weighted average of the task-specific objective f…
Maximum Roaming Multi-Task Learning
Lucas Pascal, Pietro Michiardi, Xavier Bost +2
Multi-task learning has gained popularity due to the advantages it provides with respect to resource usage and performance. Nonetheless, the joint optimization of parameters with r…
Semantic and Visual Similarities for Efficient Knowledge Transfer in CNN Training
Lucas Pascal, Xavier Bost, Benoît Huet
In recent years, representation learning approaches have disrupted many multimedia computing tasks. Among those approaches, deep convolutional neural networks (CNNs) have notably r…
The MeMAD Submission to the WMT18 Multimodal Translation Task
Stig-Arne Grönroos, Benoit Huet, Mikko Kurimo +8
This paper describes the MeMAD project entry to the WMT Multimodal Machine Translation Shared Task. We propose adapting the Transformer neural machine translation (NMT) architectur…