Robustness of Classifiers to Universal Perturbations: A Geometric Perspective

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

Research in artificial intelligence (AI) and neural network (NN) technologies has been boosted by recent hardware advances, while the emergence of self-synthesized materials such as nanocarbons has attracted much attention to overcome the limitations of Si ...

This special issue explores the burgeoning field of deep learning for remote sensing time series analysis. The 20 contributed papers showcase diverse applications, including land cover mapping, change detection, atmospheric and biophysical/biochemical para ...

Understanding how convolutional neural networks (CNNs) can efficiently learn high-dimensional functions remains a fundamental challenge. A popular belief is that these models harness the local and hierarchical structure of natural data such as images. Yet, ...

The present work proposes a framework for nonlinear model order reduction based on a Graph Convolutional Autoencoder (GCA-ROM). In the reduced order modeling (ROM) context, one is interested in obtaining real -time and many-query evaluations of parametric ...

Differentiable rendering (DR) is an emerging field that facilitates the calculation and propagation of gradients of 3D objects through images. In contrast to deep neural networks (DNNs), which generally lack the understanding of 3D objects that form the im ...

In this thesis, we study two closely related directions: robustness and generalization in modern deep learning. Deep learning models based on empirical risk minimization are known to be often non-robust to small, worst-case perturbations known as adversari ...

Interacting systems are ubiquitous in nature and engineering, ranging from particle dynamics in physics to functionally connected brain regions. Revealing interaction laws is of fundamental importance but also particularly challenging due to underlying con ...

Whereas the ability of deep networks to produce useful predictions on many kinds of data has been amply demonstrated, estimating the reliability of these predictions remains challenging. Sampling approaches such as MC-Dropout and Deep Ensembles have emerge ...

Human babies have a natural desire to interact with new toys and objects, through which they learn how the world around them works, e.g., that glass shatters when dropped, but a rubber ball does not. When their predictions are proven incorrect, such as whe ...

In this thesis we explore the applications of projective geometry, a mathematical theory of the relation between 3D scenes and their 2D images, in modern learning-based computer vision systems. This is an interesting research question which contradicts the ...

Curvilinear structures are frequently observed in a variety of domains and are essential for comprehending neural circuits, detecting fractures in materials, and determining road and irrigation canal networks. It can be costly and time-consuming to manuall ...

The successes of deep learning for semantic segmentation can in be, in part, attributed to its scale: a notion that encapsulates the largeness of these computational architectures and the labeled datasets they are trained on. These resource requirements hi ...

The goal of this paper is to characterize function distributions that general neural networks trained by descent algorithms (GD/SGD), can or cannot learn in polytime. The results are: (1) The paradigm of general neural networks trained by SGD is poly-time ...

Stereo confidence estimation aims to estimate the reliability of the estimated disparity by stereo matching. Different from the previous methods that exploit the limited input modality, we present a novel method that estimates confidence map of an initial ...

Touchscreens are nowadays the preferred choice for user interfaces in consumer electronics. Significant technological advances have been made in terms of touch sensing and visual quality. However, the haptic feedback offered by commercial products is still ...

This paper provides a theoretical study of deep neural function approximation in reinforcement learning (RL) with the ϵ-greedy exploration under the online setting. This problem setting is motivated by the successful deep Q-networks (DQN) framework that fa ...

Deep image denoisers achieve state-of-the-art results but with a hidden cost. As witnessed in recent literature, these deep networks are capable of overfitting their training distributions, causing inaccurate hallucinations to be added to the output and ge ...

Author summaryIn recent years, the application of deep learning represented a breakthrough in the mass spectrometry (MS) field by improving the assignment of the correct sequence of amino acids from observable MS spectra without prior knowledge, also known ...

State-of-the-art methods for counting people in crowded scenes rely on deep networks to estimate crowd density. While effective, these data-driven approaches rely on large amount of data annotation to achieve good performance, which stops these models from ...

Background and objectives: Parameter estimation and uncertainty quantification are crucial in computa-tional cardiology, as they enable the construction of digital twins that faithfully replicate the behavior of physical patients. Many model parameters reg ...

Robustness of Classifiers to Universal Perturbations: A Geometric Perspective

Chat with Graph Search

Split weight distribution learning for binary neural networks exploiting nano-memristive graphene/sumanene/graphene devices

Preface: Advancing deep learning for remote sensing time series data analysis

What can be learnt with wide convolutional neural networks?

A graph convolutional autoencoder approach to model order reduction for parametrized PDEs

Inverse Rendering for PET Image Reconstruction

Understanding generalization and robustness in modern deep learning

Collective relational inference for learning heterogeneous interactions

ZigZag: Universal Sampling-free Uncertainty Estimation Through Two-Step Inference

Seeking the new, learning from the unexpected: Computational models of surprise and novelty in the brain

Incorporating Projective Geometry into Deep Learning

Topologically Better Delineation of Curvilinear Structures

Confidence Matters: Applications to Semantic Segmentation

Polynomial-time universality and limitations of deep learning

Stereo Confidence Estimation via Locally Adaptive Fusion and Knowledge Distillation

Deep Learning for Localized-Haptic Feedback in Tactile Surfaces

Understanding Deep Neural Function Approximation in Reinforcement Learning via ϵ-Greedy Exploration

Image Denoising with Control over Deep Network Hallucination

Multienzyme deep learning models improve peptide de novo sequencing by mass spectrometry proteomics

Leveraging Self-Supervision for Cross-Domain Crowd Counting

Fast and robust parameter estimation with uncertainty quantification for the cardiac function