Publications related to Robust Training and Verification of Deep Neural Networks

Mutation-Attention (MuAt): deep representation learning of somatic mutations for tumour typing and subtyping

Background: Cancer genome sequencing enables accurate classification of tumours and tumour subtypes. However, prediction performance is still limited using exome-only sequencing and for tumour types with low somatic mutation burden such as many paediatri ...

2023

NMDA-driven dendritic modulation enables multitask representation learning in hierarchical sensory processing pathways

Willem Anna Mark Wybo, Bernd Albert Illing, Matthias Chinyen Tsai, Walter Senn

While sensory representations in the brain depend on context, it remains unclear how such modulations are implemented at the biophysical level, and how processing layers further in the hierarchy can extract useful features for each possible contex-tual sta ...

Washington2023

The future of human-centric eXplainable Artificial Intelligence (XAI) is not post-hoc explanations

Vinitra Swamy, Jibril Albachir Frej

Explainable Artificial Intelligence (XAI) plays a crucial role in enabling human understanding and trust in deep learning systems, often defined as determining which features are most important to a model's prediction. As models get larger, more ubiquitous ...

2023

Neural ADMIXTURE for rapid genomic clustering

Albert Dominguez Mantes

Characterizing the genetic structure of large cohorts has become increasingly important as genetic studies extend to massive, increasingly diverse biobanks. Popular methods decompose individual genomes into fractional cluster assignments with each cluster ...

SPRINGERNATURE2023

The inductive bias of deep learning: Connecting weights and functions

Guillermo Ortiz Jimenez

Years of a fierce competition have naturally selected the fittest deep learning algorithms. Yet, although these models work well in practice, we still lack a proper characterization of why they do so. This poses serious questions about the robustness, trus ...

EPFL2023

Towards Stable and Efficient Adversarial Training against $l_1$ Bounded Adversarial Attacks

Sabine Süsstrunk, Mathieu Salzmann, Yulun Jiang, Chen Liu, Zhuoyi Huang

We address the problem of stably and efficiently training a deep neural network robust to adversarial perturbations bounded by an

l_1

norm. We demonstrate that achieving robustness against

l_1

-bounded perturbations is more challenging than in the

l_2

...

2023

Euclid: Identification of asteroid streaks in simulated images using deep learning

Frédéric Courbin, Georges Meylan, Yi Wang, Richard Massey

The material composition of asteroids is an essential piece of knowledge in the quest to understand the formation and evolution of the Solar System. Visual to near-infrared spectra or multiband photometry is required to constrain the material composition o ...

Les Ulis Cedex A2023

Deep Generative Models for Autonomous Driving: from Motion Forecasting to Realistic Image Synthesis

Saeed Saadatnejad

Forecasting is a capability inherent in humans when navigating. Humans routinely plan their paths, considering the potential future movements of those around them. Similarly, to achieve comparable sophistication and safety, autonomous systems must embrace ...

EPFL2023

Source-Free Open-Set Domain Adaptation for Histopathological Images via Distilling Self-Supervised Vision Transformer

Jean-Philippe Thiran, Guillaume Marc Georges Vray, Devavrat Tomar

There is a strong incentive to develop computational pathology models to i) ease the burden of tissue typology annotation from whole slide histological images; ii) transfer knowledge, e.g., tissue class separability from the withheld source domain to the d ...

2023

Trusting the Explainers: Teacher Validation of Explainable Artificial Intelligence for Course Design

Vinitra Swamy, Mirko Marras, Sijia Du

Deep learning models for learning analytics have become increasingly popular over the last few years; however, these approaches are still not widely adopted in real-world settings, likely due to a lack of trust and transparency. In this paper, we tackle th ...

2023

Robust Training and Verification of Deep Neural Networks

Graph Chatbot

Chat with Graph Search

Mutation-Attention (MuAt): deep representation learning of somatic mutations for tumour typing and subtyping

NMDA-driven dendritic modulation enables multitask representation learning in hierarchical sensory processing pathways

The future of human-centric eXplainable Artificial Intelligence (XAI) is not post-hoc explanations

Neural ADMIXTURE for rapid genomic clustering

The inductive bias of deep learning: Connecting weights and functions

Towards Stable and Efficient Adversarial Training against $l_1$ Bounded Adversarial Attacks

Euclid: Identification of asteroid streaks in simulated images using deep learning

Deep Generative Models for Autonomous Driving: from Motion Forecasting to Realistic Image Synthesis

Source-Free Open-Set Domain Adaptation for Histopathological Images via Distilling Self-Supervised Vision Transformer

Trusting the Explainers: Teacher Validation of Explainable Artificial Intelligence for Course Design

NMDA-driven dendritic modulation enables multitask representation learning in hierarchical sensory processing pathways

Deep Generative Models for Autonomous Driving: from Motion Forecasting to Realistic Image Synthesis

Euclid: Identification of asteroid streaks in simulated images using deep learning

Mutation-Attention (MuAt): deep representation learning of somatic mutations for tumour typing and subtyping

Neural ADMIXTURE for rapid genomic clustering

The future of human-centric eXplainable Artificial Intelligence (XAI) is not post-hoc explanations

Trusting the Explainers: Teacher Validation of Explainable Artificial Intelligence for Course Design

The inductive bias of deep learning: Connecting weights and functions

Source-Free Open-Set Domain Adaptation for Histopathological Images via Distilling Self-Supervised Vision Transformer

Towards Stable and Efficient Adversarial Training against $l_1$ Bounded Adversarial Attacks