no code implementations • 29 Nov 2023 • Tristan Gomez, Harold Mouchère
Deep neural networks, while powerful for image classification, often operate as "black boxes," complicating the understanding of their decision-making processes.
no code implementations • International Workshop on Historical Document Imaging and Processing 2023 • Solène Tarride, Tristan Faine, Mélodie Boillet, Harold Mouchère, Christopher Kermorvant
However, selecting training samples based on the degree of agreement between annotators introduces a bias in the training data and does not improve the results.
Ranked #1 on Handwritten Text Recognition on Belfort
1 code implementation • 18 Jan 2023 • Jérémy Chopin, Jean-Baptiste Fasquel, Harold Mouchère, Rozenn Dahyot, Isabelle Bloch
On FASSEG data, results show that our module improves accuracy of the CNN by about 6. 3% (the Hausdorff distance decreases from 22. 11 to 20. 71).
no code implementations • 13 May 2022 • Tristan Gomez, Thomas Fréour, Harold Mouchère
An important limitation to the development of AI-based solutions for In Vitro Fertilization (IVF) is the black-box nature of most state-of-the-art models, due to the complexity of deep learning architectures, which raises potential bias and fairness issues.
no code implementations • 1 Mar 2022 • Tristan Gomez, Magalie Feyeux, Nicolas Normand, Laurent David, Perrine Paul-Gilloteaux, Thomas Fréour, Harold Mouchère
An important limitation to the development of Artificial Intelligence (AI)-based solutions for In Vitro Fertilization (IVF) is the absence of a public reference benchmark to train and evaluate deep learning (DL) models.
no code implementations • 31 Jan 2022 • Tristan Gomez, Thomas Fréour, Harold Mouchère
Due to the black-box nature of deep learning models, there is a recent development of solutions for visual explanations of CNNs.
1 code implementation • 4 Jun 2021 • Tristan Gomez, Suiyi Ling, Thomas Fréour, Harold Mouchère
The prevalence of employing attention mechanisms has brought along concerns on the interpretability of attention distributions.
no code implementations • 19 Sep 2017 • Frank Julca-Aguilar, Harold Mouchère, Christian Viard-Gaudin, Nina S. T. Hirata
We then model the recognition problem as a graph parsing problem: given an input stroke set, we search for a parse tree that represents the best interpretation of the input.