Luc Etienne Pommé

About Me

Starting from 1st December 2024, I am a postdoc at the University of Montpellier in France. I am working on new efficient techniques to visualize and analyze spatio-temporal data with the help of deep learning and explainability approaches, in the LIRMM (Laboratoire d'Informatique, de Robotique et de Microélectronique de Montpellier) institute, in France. I am a member of the ADVANSE (``ADVanced Analytics for data SciencE'') team.

I obtained my PhD in computer science at the LaBRI (Laboratoire Bordelais de Recherche en Informatique), in Bordeaux, in 2024 (supervisor: David AUBER). The topic of my PhD thesis is the ``Visualization for the Explanation of Deep Neural Networks''.

I obtained my Master Degree in Computer Sciences (Image and Sound processing) in 2020 at the University of Bordeaux.

I was awarded the Isabelle Attali prize in 2018.

---

Research

Keywords

Information Visualization, Deep Learning, XAI (eXplainable Artificial Intelligence), Explainable pruning, Computer Vision

Interests

The understanding of the world that surrounds us requires the observation and analysis of the phenomena that govern it. The observation of these phenomena requires the systematic collection of large quantities of data to build models representative of the reality. Eventually, the models are used to extract knowledge and properties about the object of study.

The more complex the data, the more difficult it becomes for a human to find processing rules or concepts to extract. This is where AI (e.g. machine / deep learning) algorithms come in to help. The superior performance of these algorithms being no longer in question, they are widely used in matters where decision making is required.

In addition to the worrying ecological footprint of these tools, the ethical aspect is also at stake. Indeed, these algorithms are often considered as ``black boxes'', making arbitrary decisions without justification. The question of responsability, in case of bad decision arises. It becomes necessary to build systems enabling to understand the behavior (either good or bad) of such algorithms in order to improve the trust we can put in their decisions. This problem has given rise to a field of research called XAI (eXplainable AI), which is currently in full expansion.

Within this research field, I am particularly interested in building new visualizations to highlight AIs behavior, and to try to provide justifications to their decisions. On the other hand, I am interested in tackling the over-parameterization problem of neural networks. Considering neural networks as directed graphs (where nodes are parameters/filters/neurons, and edges are input/output data), my work consists in studying the role of graph nodes and remove those that do not have a significant importance in the decisions. The induced sub-graphs represent sub-networks that achieve comparable performance on a task, with significantly less parameters, operations required to process input data, computation resources, and thus significantly less energy consumption.

---

Publications

2024

• Luc Etienne Pommé, Romain Bourqui, Romain Giot, David Auber ``Relative Confusion Matrix: An Efficient Visualization for the Comparison of Classification Models'', Artificial Intelligence and Visualization: Advancing Visual Knowledge Discovery, 1126, Springer Nature Switzerland, pp.223-243, Studies in Computational Intelligence, 2024 [10.1007/978-3-031-46549-9_7] RCM implementation

2023

• Luc Etienne Pommé, Romain Bourqui, Romain Giot ``H²O: Heatmap by Hierarchical Occlusion'', CBMI 2023, April 2023 Orléans, France [10.1145/3617233.3617256] H²O implementation

• Luc Etienne Pommé, Romain Bourqui, Romain Giot, Jason Vallet, David Auber ``NetPrune: A sparklines visualization for network pruning'', Visual Informatics, May 2023 [10.1016/j.visinf.2023.04.001] NetPrune implementation

2022

• Luc Etienne Pommé, Romain Bourqui, Romain Giot, David Auber ``Relative Confusion Matrix: Efficient Comparison of Decision Models'', IV2022-IVE-Information Visualization Evaluation & User Studies, Jul 2022, Vienne, Austria [10.1109/IV56949.2022.00025] RCM implementation

---

Teaching

Master

• Analysis, Classification, Indexing of Data (ACID) x3, 1st year of Computer Science Master, University of Bordeaux
• Artificial Intelligence, 1st year of Computer Science Master, University of Bordeaux

---

Licence

• Algorithmics for elementary data structures, 2nd year of Computer Science Licence, University of Bordeaux
• Introduction to software development, 1st year of Computer Science Licence, University of Bordeaux
• Technological projects, 2nd year of Computer Science Licence, University of Bordeaux

UTI (Universitary Technical Institute)

• Introduction to Programming, 1st year of UTI, University of Bordeaux
• Discrete Mathematics, 1st year of UTI, University of Bordeaux
• Quality in Programming, 1st year of UTI, University of Bordeaux
• Programming Project (Opening and personalization activities), 2nd year of UTI, University of Bordeaux

---

Project supervision

• Semester Project, 3rd year of Engineering School option AI, ENSEIRB Bordeaux
• Semester Project, 1st year of Computer Science Master, University of Bordeaux
• Graduation Project x2, 2nd year of Software Engineering Master, University of Bordeaux
• Technological Project, 2nd year of Computer Science Licence, University of Bordeaux

---

Resources

[ 0000-0002-4690-7260 ]

[ luc-etienne-pomme ]

[ Luc-Etienne POMMÉ ]

---