Stéphane Ayache
YOU?
Author Swipe
View article: AI-Driven Plasma Denaturation Profiling for Multi-Cancer Detection
AI-Driven Plasma Denaturation Profiling for Multi-Cancer Detection Open
Many cancers cannot be detected early due to lack of effective disease biomarkers, leading to poor prognosis. We applied an existing biophysical technology nanoDSF in a novel way to answer this unmet biomedical need. We developed a breakth…
View article: Transcanal Endoscopic Stapes Surgery
Transcanal Endoscopic Stapes Surgery Open
Stapes surgery is traditionally based on microscopic approach. For several years, endoscopy developed from a combined approach as a complementary device to microscope to an exclusive transcanal approach in various fields of ear surgery. Th…
View article: Transcanal Endoscopic Ear Surgery for Cholesteatoma: Royal Belgian Society of Otorhinolaryngology
Transcanal Endoscopic Ear Surgery for Cholesteatoma: Royal Belgian Society of Otorhinolaryngology Open
Surgery for cholesteatoma is traditionally based on microscopic approach. For several years, endoscopy developed from a combined approach as a complementary device to microscope to an exclusive transcanal approach in various fields of ear …
View article: Mitigation of gender bias in automatic facial non-verbal behaviors generation
Mitigation of gender bias in automatic facial non-verbal behaviors generation Open
Research on non-verbal behavior generation for social interactive agents\nfocuses mainly on the believability and synchronization of non-verbal cues with\nspeech. However, existing models, predominantly based on deep learning\narchitecture…
View article: Mitigation of gender bias in automatic facial non-verbal behaviors generation
Mitigation of gender bias in automatic facial non-verbal behaviors generation Open
Research on non-verbal behavior generation for social interactive agents focuses mainly on the believability and synchronization of non-verbal cues with speech. However, existing models, predominantly based on deep learning architectures, …
View article: Analyzing gender bias in the non-verbal behaviors of generative systems
Analyzing gender bias in the non-verbal behaviors of generative systems Open
International audience
View article: Opti-CAM: Optimizing saliency maps for interpretability
Opti-CAM: Optimizing saliency maps for interpretability Open
International audience
View article: Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings
Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings Open
The cerebral processing of voice information is known to engage, in human as well as non-human primates, “temporal voice areas” (TVAs) that respond preferentially to conspecific vocalizations. However, how voice information is represented …
View article: Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings
Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings Open
The cerebral processing of voice information is known to engage, in human as well as non-human primates, “temporal voice areas” (TVAs) that respond preferentially to conspecific vocalizations. However, how voice information is represented …
View article: Author response: Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings
Author response: Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings Open
The cerebral processing of voice information is known to engage, in human as well as non-human primates, "temporal voice areas" (TVAs) that respond preferentially to conspecific vocalizations. However, how voice information is represented …
View article: Reviewer #3 (Public Review): Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings
Reviewer #3 (Public Review): Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings Open
The cerebral processing of voice information is known to engage, in human as well as non-human primates, "temporal voice areas" (TVAs) that respond preferentially to conspecific vocalizations. However, how voice information is represented …
View article: Implicit Regularization in Deep Tucker Factorization: Low-Rankness via Structured Sparsity
Implicit Regularization in Deep Tucker Factorization: Low-Rankness via Structured Sparsity Open
International audience
View article: CA-Stream: Attention-based pooling for interpretable image recognition
CA-Stream: Attention-based pooling for interpretable image recognition Open
Explanations obtained from transformer-based architectures in the form of raw attention, can be seen as a class-agnostic saliency map. Additionally, attention-based pooling serves as a form of masking the in feature space. Motivated by thi…
View article: A Learning Paradigm for Interpretable Gradients
A Learning Paradigm for Interpretable Gradients Open
This paper studies interpretability of convolutional networks by means of saliency maps. Most approaches based on Class Activation Maps (CAM) combine information from fully connected layers and gradient through variants of backpropagation.…
View article: Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings
Reconstructing Voice Identity from Noninvasive Auditory Cortex Recordings Open
The cerebral processing of voice information is known to engage, in human as well as non-human primates, “temporal voice areas” (TVAs) that respond preferentially to conspecific vocalizations. However, how voice information is represented …
View article: A Learning Paradigm for Interpretable Gradients
A Learning Paradigm for Interpretable Gradients Open
This paper studies interpretability of convolutional networks by means of\nsaliency maps. Most approaches based on Class Activation Maps (CAM) combine\ninformation from fully connected layers and gradient through variants of\nbackpropagati…
View article: Towards the generation of synchronized and believable non-verbal facial behaviors of a talking virtual agent
Towards the generation of synchronized and believable non-verbal facial behaviors of a talking virtual agent Open
International audience
View article: Towards the generation of synchronized and believable non-verbal facial behaviors of a talking virtual agent
Towards the generation of synchronized and believable non-verbal facial behaviors of a talking virtual agent Open
This paper introduces a new model to generate rhythmically relevant non-verbal facial behaviors for virtual agents while they speak. The model demonstrates perceived performance comparable to behaviors directly extracted from the data and …
View article: DP-Net: Learning Discriminative Parts for Image Recognition
DP-Net: Learning Discriminative Parts for Image Recognition Open
This paper presents Discriminative Part Network (DP-Net), a deep architecture with strong interpretation capabilities, which exploits a pretrained Convolutional Neural Network (CNN) combined with a part-based recognition module. This syste…
View article: Plasma nanoDSF Denaturation Profile at Baseline Is Predictive of Glioblastoma EGFR Status
Plasma nanoDSF Denaturation Profile at Baseline Is Predictive of Glioblastoma EGFR Status Open
Glioblastoma (GBM) is the most frequent and aggressive primary brain tumor in adults. Recently, we demonstrated that plasma denaturation profiles of glioblastoma patients obtained using Differential Scanning Fluorimetry can be automaticall…
View article: Opti-CAM: Optimizing saliency maps for interpretability
Opti-CAM: Optimizing saliency maps for interpretability Open
Methods based on class activation maps (CAM) provide a simple mechanism to interpret predictions of convolutional neural networks by using linear combinations of feature maps as saliency maps. By contrast, masking-based methods optimize a …
View article: Automatic facial expressions, gaze direction and head movements generation of a virtual agent
Automatic facial expressions, gaze direction and head movements generation of a virtual agent Open
In this article, we present two models to jointly and automatically generate the head, facial and gaze movements of a virtual agent from acoustic speech features. Two architectures are explored: a Generative Adversarial Network and an Adve…
View article: Do Vision-and-Language Transformers Learn Grounded Predicate-Noun Dependencies?
Do Vision-and-Language Transformers Learn Grounded Predicate-Noun Dependencies? Open
Recent advances in vision-and-language modeling have seen the development of Transformer architectures that achieve remarkable performance on multimodal reasoning tasks. Yet, the exact capabilities of these black-box models are still poorl…
View article: Implicit Regularization with Polynomial Growth in Deep Tensor Factorization
Implicit Regularization with Polynomial Growth in Deep Tensor Factorization Open
We study the implicit regularization effects of deep learning in tensor factorization. While implicit regularization in deep matrix and 'shallow' tensor factorization via linear and certain type of non-linear neural networks promotes low-r…
View article: Implicit Regularization with Polynomial Growth in Deep Tensor Factorization
Implicit Regularization with Polynomial Growth in Deep Tensor Factorization Open
International audience
View article: Are Vision-Language Transformers Learning Multimodal Representations? A Probing Perspective
Are Vision-Language Transformers Learning Multimodal Representations? A Probing Perspective Open
In recent years, joint text-image embeddings have significantly improved thanks to the development of transformer-based Vision-Language models. Despite these advances, we still need to better understand the representations produced by thos…
View article: Are Vision-Language Transformers Learning Multimodal Representations? A Probing Perspective
Are Vision-Language Transformers Learning Multimodal Representations? A Probing Perspective Open
International audience