Neural Networks for Information Retrieval

Machine learning plays an important role in many aspects of modern IR systems, and deep learning is applied to all of those. The fast pace of modern-day research into deep learning has given rise to many different approaches to many different IR problems.

  Download slides (PDF, 18MB)

Full-day tutorial at SIGIR 2017 on August 7, 2017

Table of contents

The tutorial is organized in 8 parts as described below. For more details, see the overview paper on arXiv.

Time slot Topic Presenters
09:00 - 09:05 Introduction Tom Kenter section slides
09:05 - 10:00 Preliminaries Tom Kenter, Maarten de Rijke section slides
10:00 - 10:30 Text Matching I Alexey Borisov, Bhaskar Mitra section slides
10:30 - 10:50 Coffee break
10:50 - 11:20 Text Matching I (cont'd) Alexey Borisov, Bhaskar Mitra section slides
11:20 - 12:20 Text Matching II Mostafa Deghani, Christophe Van Gysel section slides
12:20 - 14:00 Lunch break
14:00 - 14:45 Learning to Rank Christophe Van Gysel, Bhaskar Mitra section slides
14:45 - 15:30 Modeling User Behavior Alexey Borisov, Maarten de Rijke section slides
15:30 - 15:50 Coffee break
15:50 - 16:35 Generating Responses Mostafa Deghani, Tom Kenter section slides
16:35 - 17:20 Wrap Up All presenters section slides

ReferencesAll slides

Citation

If you wish to refer to the tutorial in your scientific publication, please refer to our overview paper:

@inproceedings{Kenter2017nn4ir,
  Author = {Kenter, Tom and Borisov, Alexey and Van Gysel, Christophe and Dehghani, Mostafa and de Rijke, Maarten and Mitra, Bhaskar},
  Booktitle = {SIGIR 2017},
  Organization = {ACM},
  Pages = {1403--1406},
  Title = {Neural Networks for Information Retrieval},
  Year = {2017}}

Organizers