Learning the Underlying Structure of NLP Tasks
Event details
| Date | 09.07.2019 |
| Hour | 10:00 › 12:00 |
| Speaker | Jean-Baptiste Cordonnier |
| Location | |
| Category | Conferences - Seminars |
EDIC candidacy exam
Exam president: Dr. François Fleuret
Thesis advisor: Prof. Martin Jaggi
Co-examiner: Prof. Robert West
Abstract
Progress in Natural Language Processing has been
driven by the quest of architectures capturing the structure of
text and more recently by novel pre-training tasks on large
text corpora. In this report, we discuss two important papers
that shifted the NLP researchers’ attention toward better semisupervised
tasks and downstream training using Multi-Task
Learning. We then step back and study how the Computer Vision
community studies relationships between visual tasks and which
lessons apply to NLP. I finally outline my research proposal for
the rest of the doctoral studies.
Background papers
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, by Devlin,J., et al.
Multi-Task Deep Neural Networks for Natural Language Understanding, by Liu, X., et al.
Taskonomy: Disentangling Task Transfer Learning, by Zamir, A., et al.
Exam president: Dr. François Fleuret
Thesis advisor: Prof. Martin Jaggi
Co-examiner: Prof. Robert West
Abstract
Progress in Natural Language Processing has been
driven by the quest of architectures capturing the structure of
text and more recently by novel pre-training tasks on large
text corpora. In this report, we discuss two important papers
that shifted the NLP researchers’ attention toward better semisupervised
tasks and downstream training using Multi-Task
Learning. We then step back and study how the Computer Vision
community studies relationships between visual tasks and which
lessons apply to NLP. I finally outline my research proposal for
the rest of the doctoral studies.
Background papers
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, by Devlin,J., et al.
Multi-Task Deep Neural Networks for Natural Language Understanding, by Liu, X., et al.
Taskonomy: Disentangling Task Transfer Learning, by Zamir, A., et al.
Practical information
- General public
- Free