FonMTL: Towards Multitask Learning for the Fon Language

Bonaventure F. P. Dossou, Iffanice Houndayi, Pamely Joanis Zantou, Gilles HACHEME

Keywords: multitask learning, african nlp

TL;DR: Multitask Learning for the Fon Language

Abstract: This paper presents the first explorative approach to multitask learning, for model capabilities enhancement in Natural Language Processing for the Fon language. Specifically, we explore the tasks of Named Entity Recognition (NER) and Part of Speech Tagging (POS) for Fon. We leverage two language model heads as encoders to build shared representations for the inputs, and we use linear layers blocks for classification relative to each task. Our results on the NER task for Fon show a competitive performance compared to the performance of individual several multilingual models. Additionally, we perform a few ablation studies to leverage the efficiency of two different loss combination strategies and find out that the equal loss weighting approach works best in our case.

Submission Category: Machine learning algorithms

Submission Number: 8