Skip to Main content Skip to Navigation
Conference papers

Continual Learning of Long Topic Sequences in Neural Information Retrieval

Thomas Gerald 1 Laure Soulier 1 
1 MLIA - Machine Learning and Information Access
ISIR - Institut des Systèmes Intelligents et de Robotique
Abstract : In information retrieval (IR) systems, trends and users' interests may change over time, altering either the distribution of requests or contents to be recommended. Since neural ranking approaches heavily depend on the training data, it is crucial to understand the transfer capacity of recent IR approaches to address new domains in the long term. In this paper, we first propose a dataset based upon the MSMarco corpus aiming at modeling a long stream of topics as well as IR property-driven controlled settings. We then in-depth analyze the ability of recent neural IR models while continually learning those streams. Our empirical study highlights in which particular cases catastrophic forgetting occurs (e.g., level of similarity between tasks, peculiarities on text length, and ways of learning models) to provide future directions in terms of model design.
Complete list of metadata

https://hal.sorbonne-universite.fr/hal-03563308
Contributor : Laure Soulier Connect in order to contact the contributor
Submitted on : Wednesday, February 9, 2022 - 3:57:16 PM
Last modification on : Monday, June 27, 2022 - 12:30:33 PM
Long-term archiving on: : Tuesday, May 10, 2022 - 7:04:09 PM

File

ecir2022_ContinualRanking_cr (...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03563308, version 1

Citation

Thomas Gerald, Laure Soulier. Continual Learning of Long Topic Sequences in Neural Information Retrieval. 44th European Conference on Information Retrieval (ECIR 2022), Apr 2022, Stavanger, Norway. ⟨hal-03563308⟩

Share

Metrics

Record views

9

Files downloads

12