ComplexNeuroViz (2021)

June-August 2021

ComplexNeuroViz : Complexity Visualisation for Neural Machine Translation

Funding

• 3,500 euros for trainees

The ComplexNeuroViz is funded by 2020 diiP Masters Internship Project

PI

• Nicolas Ballier

Joint research with 

• Guillaume Wisniewski (LLF)
• Jean-Baptiste Yunès (IRIF)

Resume

A Master student Internship Project on data science, in connection to interdisciplinary collaborations between translation studies, computational linguistics and computations science for two trainees.

Trainees

• Bilal FAYE (MSc in Machine Learning for Data Science at UP)
• Zong-You KE (MSc in Computational linguistics at UP)

Aim

This project aims at improving visualisation of neural networks when they process linguistic data in the context of neural translation. The analysis of neural networks with visualization toolkits mainly focused on the activation of layers and neurons for classification tasks. Compared to the two state-of-the-art visualization toolkits for neural translation, our major contribution consists in adding one currently neglected set of features in Feature Visualization: linguistic complexity. The aim of this project is to adapt current neural visualization toolkits to visualize and diagnose attention weights and linguistic complexity metrics in order to analyse the effects of the complexity of the linguistic input at token level, at phrase level and at sentence level.

Systems

tbc

Deliverables

tbc