Publications associées à Large deviations in the perceptron model and consequences for active learning

Publications associées (24)

Do You Trust Your Model? Emerging Malware Threats in the Deep Learning Ecosystem

Training high-quality deep learning models is a challenging task due to computational and technical requirements. A growing number of individuals, institutions, and companies increasingly rely on pre-trained, third-party models made available in public rep ...

2025

Adversarial Training For Defense Against Label Poisoning Attacks

Volkan Cevher

As machine learning models grow in complexity and increasingly rely on publicly sourced data, such as the human-annotated labels used in training large language models, they become more vulnerable to label poisoning attacks. These attacks, in which adversa ...

2025

Controlled Training Data Generation with Diffusion Models

Aleksandr Alekseev, Shuqing Teresa Yeo, Andrei Atanov

We present a method to control a text-to-image generative model to produce training data useful for supervised learning. Unlike previous works that employ an open-loop approach via pre-defined prompts to generate new data using either a language model or h ...

2025

Social-Pose: Enhancing Trajectory Prediction with Human Body Pose

Alexandre Massoud Alahi, Saeed Saadatnejad, Yang Gao

Accurate human trajectory prediction is one of the most crucial tasks for autonomous driving, ensuring its safety. Yet, existing models often fail to fully leverage the visual cues that humans subconsciously communicate when navigating the space. In this w ...

IEEE Institute of Electrical and Electronics Engineers2025

Fine-tuning protein language models to understand the functional impact of missense variants

Jacques Fellay

Elucidating the functional effects of missense variants is crucial yet challenging. To investigate their impact, we fine-tuned protein language models, including ESM2 and ProtT5, to classify 20 protein features at amino acid resolution. In addition, we tra ...

2025

Sim-to-Real Causal Transfer: A Metric Learning Approach to Causally-Aware Interaction Representations

Alexandre Massoud Alahi, Yuejiang Liu, Po-Chien Luan, Ahmad Rahimi

Modeling spatial-temporal interactions among neighboring agents is at the heart of multi-agent problems such as motion forecasting and crowd navigation. Despite notable progress, it remains unclear to which extent modern representations can capture the cau ...

2025

Uncertainty-aware diabetic retinopathy detection using deep learning enhanced by Bayesian approaches

Zaffar Ahmed Shaikh

Deep learning-based medical image analysis has shown strong potential in disease categorization, segmentation, detection, and even prediction. However, in high-stakes and complex domains like healthcare, the opaque nature of these models makes it challengi ...

2025

Machine learning Hubbard parameters with equivariant neural networks

Nicola Marzari, Iurii Timrov, Martin Uhrin, Luca Binci, Austin Jacob Zadoks

Density-functional theory with extended Hubbard functionals (DFT + U + V) provides a robust framework to accurately describe complex materials containing transition-metal or rare-earth elements. It does so by mitigating self-interaction errors inherent to ...

NATURE PORTFOLIO2025

Parameter-efficient Methods for Training and Inference of Transformer-based Models

Mohammadreza Banaei

Recent advancements in natural language processing, driven by the success of large language models, have led to the scaling up of models to unprecedented sizes. While these models have achieved remarkable performance across various tasks, their computation ...

EPFL2025

TRACKASTRA: Transformer-Based Cell Tracking for Live-Cell Microscopy

Martin Weigert, Benjamin Tobias Gallusser

Cell tracking is a ubiquitous image analysis task in live-cell microscopy. Unlike multiple object tracking (MOT) for natural images, cell tracking typically involves hundreds of similar-looking objects that can divide in each frame, making it a particularl ...

Springer Science and Business Media Deutschland GmbH2025

Helmet material design for mitigating traumatic axonal injuries through AI-driven constitutive law enhancement

Dominique Pioletti, Pierre-Etienne Bourban, Naser Nasrollahzadeh Mamaghani, Vincent Marc Varanges, Pezhman Eghbalishamsabadi

Sports helmets provide incomplete protection against brain injuries. Here we aim to improve helmet liner efficiency by employing a novel approach that optimizes their properties. By exploiting a finite element model that simulates head impacts, we develope ...

2025

Towards Reliable Evaluation and Fast Training of Robust Semantic Segmentation Models

Francesco Croce

Adversarial robustness has been studied extensively in image classification, especially for the ℓ∞-threat model, but significantly less so for related tasks such as object detection and semantic segmentation, where attacks turn out to be a much harder opti ...

Springer Science and Business Media Deutschland GmbH2025

Joint Multiband Deconvolution for Euclid and Vera C. Rubin Images

Frédéric Courbin, Pascale Jablonka, Utsav Akhaury

With the advent of surveys like Euclid and Vera C. Rubin, astrophysicists will have access to both deep, high-resolution images and multiband images. However, these two types are not simultaneously available in any single dataset. It is therefore vital to ...

EDP Sciences SA2025

Songdo Traffic: High Accuracy Georeferenced Vehicle Trajectories from a Large-Scale Study in a Smart City

Nikolaos Geroliminis, Robert Fonod

The Songdo Traffic dataset delivers precisely georeferenced vehicle trajectories captured through high-altitude bird's-eye view (BeV) drone footage over Songdo International Business District, South Korea. Comprising approximately 700,000 unique trajectori ...

Zenodo2025

Automatic Parkinson's disease detection from speech: Layer selection vs adaptation of foundation models

Tilak Purohit

In this work, we investigate Speech Foundation Models (SFMs) for Parkinson's Disease (PD) detection. We explore two main approaches: (1) using SFMs as frozen feature extractors and, (2) fine-tuning/adapting SFMs for PD detection. We propose a cross-validat ...

Institute of Electrical and Electronics Engineers Inc.2025

Qualify-as-you-go: sensor fusion of optical and acoustic signatures with contrastive deep learning for multi-material composition monitoring in laser powder bed fusion process

Christian Leinenbach, Antonios Nikolaos Baganis

Growing demand for multi-material Laser Powder Bed Fusion (LPBF) faces process control and quality monitoring challenges, particularly in ensuring precise material composition. This study explores optical and acoustic emission signals during LPBF processes ...

Taylor & Francis Ltd2024

A prediction rigidity formalism for low-cost uncertainties in trained neural networks

Michele Ceriotti, Federico Grasselli, Sanggyu Chong, Filippo Bigi

Quantifying the uncertainty of regression models is essential to ensure their reliability, particularly since their application often extends beyond their training domain. Based on the solution of a constrained optimization problem, this work proposes ‘pre ...

2024

Enhancing Peer Review with AI-Powered Suggestion Generation Assistance: Investigating the Design Dynamics

Xiaotian Su, Seyed Parsa Neshaei

While writing peer reviews resembles an important task in science, education, and large organizations, providing fruitful suggestions to peers is not a straightforward task, as different user interaction designs of text suggestion interfaces can have diver ...

Association for Computing Machinery2024

Chemical education in digital chemistry

Philippe Schwaller

In this digital age where machine learning has won the Nobel Prizes in both Physics and Chemistry, it is ever more important to give chemistry students an educational advantage that will enable them to use the tools of artificial intelligence and machine l ...

CELL PRESS2024

A two-level resolution neural network with enhanced interpretability for freeway traffic forecasting

Nikolaos Geroliminis, Semin Kwak, Danya Li

© The Author(s) 2024.Deep learning models are widely used for traffic forecasting on freeways due to their ability to learn complex temporal and spatial relationships. In particular, graph neural networks, which integrate graph theory into deep learning, h ...

Nature Research2024