Christian Gagné
YOU?
Author Swipe
View article: Personalized Federated Fine-Tuning of Vision Foundation Models for Healthcare
Personalized Federated Fine-Tuning of Vision Foundation Models for Healthcare Open
Foundation models open up new possibilities for the use of AI in healthcare. However, even when pre-trained on health data, they still need to be fine-tuned for specific downstream tasks. Furthermore, although foundation models reduce the …
View article: Robust Fine-Tuning from Non-Robust Pretrained Models: Mitigating Suboptimal Transfer With Adversarial Scheduling
Robust Fine-Tuning from Non-Robust Pretrained Models: Mitigating Suboptimal Transfer With Adversarial Scheduling Open
Fine-tuning pretrained models is a standard and effective workflow in modern machine learning. However, robust fine-tuning (RFT), which aims to simultaneously achieve adaptation to a downstream task and robustness to adversarial examples, …
View article: A Guide to Robust Generalization: The Impact of Architecture, Pre-training, and Optimization Strategy
A Guide to Robust Generalization: The Impact of Architecture, Pre-training, and Optimization Strategy Open
Deep learning models operating in the image domain are vulnerable to small input perturbations. For years, robustness to such perturbations was pursued by training models from scratch (i.e., with random initializations) using specialized l…
View article: A Self-Supervised Foundation Model for Robust and Generalizable Representation Learning in STED Microscopy
A Self-Supervised Foundation Model for Robust and Generalizable Representation Learning in STED Microscopy Open
Foundation Models (FMs) have dramatically increased the potential and power of deep learning algorithms in the fields of natural language processing and computer vision. However, their application in specialized fields like biomedical imag…
View article: TrackPGD: Efficient Adversarial Attack using Object Binary Masks against Robust Transformer Trackers
TrackPGD: Efficient Adversarial Attack using Object Binary Masks against Robust Transformer Trackers Open
View article: A Layer Selection Approach to Test Time Adaptation
A Layer Selection Approach to Test Time Adaptation Open
Test Time Adaptation (TTA) addresses the problem of distribution shift by adapting a pretrained model to a new domain during inference. When faced with challenging shifts, most methods collapse and perform worse than the original pretraine…
View article: Revisiting Data Augmentation for Ultrasound Images
Revisiting Data Augmentation for Ultrasound Images Open
Data augmentation is a widely used and effective technique to improve the generalization performance of deep neural networks. Yet, despite often facing limited data availability when working with medical images, it is frequently underutili…
View article: Adversarial Bounding Boxes Generation (ABBG) Attack against Visual Object Trackers
Adversarial Bounding Boxes Generation (ABBG) Attack against Visual Object Trackers Open
International audience
View article: Quantitative analysis of miniature synaptic calcium transients using positive unlabeled deep learning
Quantitative analysis of miniature synaptic calcium transients using positive unlabeled deep learning Open
We developed a positive unlabeled deep learning scheme for detection and segmentation of miniature synaptic calcium transients. Combining deep learning and feature analysis, it measures the impact of cLTP on transient morphology and dynami…
View article: TrackPGD: Efficient Adversarial Attack using Object Binary Masks against Robust Transformer Trackers
TrackPGD: Efficient Adversarial Attack using Object Binary Masks against Robust Transformer Trackers Open
Adversarial perturbations can deceive neural networks by adding small, imperceptible noise to the input. Recent object trackers with transformer backbones have shown strong performance on tracking datasets, but their adversarial robustness…
View article: Predicting the Population Risk of Suicide Using Routinely Collected Health Administrative Data in Quebec, Canada: Model-Based Synthetic Estimation Study
Predicting the Population Risk of Suicide Using Routinely Collected Health Administrative Data in Quebec, Canada: Model-Based Synthetic Estimation Study Open
Background Suicide is a significant public health issue. Many risk prediction tools have been developed to estimate an individual’s risk of suicide. Risk prediction models can go beyond individual risk assessment; one important application…
View article: Reproducibility Study on Adversarial Attacks Against Robust Transformer Trackers
Reproducibility Study on Adversarial Attacks Against Robust Transformer Trackers Open
New transformer networks have been integrated into object tracking pipelines and have demonstrated strong performance on the latest benchmarks. This paper focuses on understanding how transformer trackers behave under adversarial attacks a…
View article: Layerwise Early Stopping for Test Time Adaptation
Layerwise Early Stopping for Test Time Adaptation Open
Test Time Adaptation (TTA) addresses the problem of distribution shift by enabling pretrained models to learn new features on an unseen domain at test time. However, it poses a significant challenge to maintain a balance between learning n…
View article: Explainable artificial intelligence models for predicting risk of suicide using health administrative data in Quebec
Explainable artificial intelligence models for predicting risk of suicide using health administrative data in Quebec Open
Suicide is a complex, multidimensional event, and a significant challenge for prevention globally. Artificial intelligence (AI) and machine learning (ML) have emerged to harness large-scale datasets to enhance risk detection. In order to t…
View article: Generalizing across Temporal Domains with Koopman Operators
Generalizing across Temporal Domains with Koopman Operators Open
In the field of domain generalization, the task of constructing a predictive model capable of generalizing to a target domain without access to target data remains challenging. This problem becomes further complicated when considering evol…
View article: Analyzing Data Augmentation for Medical Images: A Case Study in Ultrasound Images
Analyzing Data Augmentation for Medical Images: A Case Study in Ultrasound Images Open
Data augmentation is one of the most effective techniques to improve the generalization performance of deep neural networks. Yet, despite often facing limited data availability in medical image analysis, it is frequently underutilized. Thi…
View article: Generalizing across Temporal Domains with Koopman Operators
Generalizing across Temporal Domains with Koopman Operators Open
In the field of domain generalization, the task of constructing a predictive model capable of generalizing to a target domain without access to target data remains challenging. This problem becomes further complicated when considering evol…
View article: Detecting Brittle Decisions for Free: Leveraging Margin Consistency in Deep Robust Classifiers
Detecting Brittle Decisions for Free: Leveraging Margin Consistency in Deep Robust Classifiers Open
Despite extensive research on adversarial training strategies to improve robustness, the decisions of even the most robust deep learning models can still be quite sensitive to imperceptible perturbations, creating serious risks when deploy…
View article: Unmixing Optical Signals from Undersampled Volumetric Measurements by Filtering the Pixel Latent Variables
Unmixing Optical Signals from Undersampled Volumetric Measurements by Filtering the Pixel Latent Variables Open
The development of signal unmixing algorithms is essential for leveraging multimodal datasets acquired through a wide array of scientific imaging technologies, including hyperspectral or time-resolved acquisitions. In experimental physics,…
View article: Hessian Aware Low-Rank Perturbation for Order-Robust Continual Learning
Hessian Aware Low-Rank Perturbation for Order-Robust Continual Learning Open
Continual learning aims to learn a series of tasks sequentially without forgetting the knowledge acquired from the previous ones. In this work, we propose the Hessian Aware Low-Rank Perturbation algorithm for continual learning. By modelin…
View article: Denoising autoencoder genetic programming: strategies to control exploration and exploitation in search
Denoising autoencoder genetic programming: strategies to control exploration and exploitation in search Open
Denoising autoencoder genetic programming (DAE-GP) is a novel neural network-based estimation of distribution genetic programming approach that uses denoising autoencoder long short-term memory networks as a probabilistic model to replace …
View article: Resolution enhancement with a task-assisted GAN to guide optical nanoscopy image analysis and acquisition
Resolution enhancement with a task-assisted GAN to guide optical nanoscopy image analysis and acquisition Open
View article: Domain Agnostic Image-to-image Translation using Low-Resolution Conditioning
Domain Agnostic Image-to-image Translation using Low-Resolution Conditioning Open
Generally, image-to-image translation (i2i) methods aim at learning mappings across domains with the assumption that the images used for translation share content (e.g., pose) but have their own domain-specific information (a.k.a. style). …
View article: Confocal and STED Live F-actin dataset
Confocal and STED Live F-actin dataset Open
Paired confocal and STED images of F-actin nanostructures in living neurons using the far-red fluorogenic dye SiR-Actin. This dataset was used to train and test the TA-GAN model for confocal-to-STED super-resolution of axonal and dendritic…
View article: Confocal and STED Live F-actin dataset
Confocal and STED Live F-actin dataset Open
Paired confocal and STED images of F-actin nanostructures in living neurons using the far-red fluorogenic dye SiR-Actin. This dataset was used to train and test the TA-GAN model for confocal-to-STED super-resolution of axonal and dendritic…
View article: Improved Robustness Against Adaptive Attacks With Ensembles and Error-Correcting Output Codes
Improved Robustness Against Adaptive Attacks With Ensembles and Error-Correcting Output Codes Open
Neural network ensembles have been studied extensively in the context of adversarial robustness and most ensemble-based approaches remain vulnerable to adaptive attacks. In this paper, we investigate the robustness of Error-Correcting Outp…
View article: A case–control study on predicting population risk of suicide using health administrative data: a research protocol
A case–control study on predicting population risk of suicide using health administrative data: a research protocol Open
Introduction Suicide has a complex aetiology and is a result of the interaction among the risk and protective factors at the individual, healthcare system and population levels. Therefore, policy and decision makers and mental health servi…
View article: On Learning Fairness and Accuracy on Multiple Subgroups
On Learning Fairness and Accuracy on Multiple Subgroups Open
We propose an analysis in fair learning that preserves the utility of the data while reducing prediction disparities under the criteria of group sufficiency. We focus on the scenario where the data contains multiple or even many subgroups,…
View article: Evolving Domain Generalization
Evolving Domain Generalization Open
Domain generalization aims to learn a predictive model from multiple different but related source tasks that can generalize well to a target task without the need of accessing any target data. Existing domain generalization methods ignore …
View article: Fair Representation Learning through Implicit Path Alignment
Fair Representation Learning through Implicit Path Alignment Open
We consider a fair representation learning perspective, where optimal predictors, on top of the data representation, are ensured to be invariant with respect to different sub-groups. Specifically, we formulate this intuition as a bi-level …