Recent years have witnessed an explosion in the development of novel
pre...
Feature visualization has gained substantial popularity, particularly af...
In recent years, concept-based approaches have emerged as some of the mo...
One of the most impactful findings in computational neuroscience over th...
Deep neural networks (DNNs) are known to have a fundamental sensitivity ...
Transformer architectures are complex and their use in NLP, while it has...
Deploying deep learning models in real-world certified systems requires ...
An important milestone for AI is the development of algorithms that can
...
We present an application of conformal prediction, a form of uncertainty...
Attribution methods are a popular class of explainability methods that u...
The many successes of deep neural networks (DNNs) over the past decade h...
Vision transformers are nowadays the de-facto preference for image
class...
We argue that, when learning a 1-Lipschitz neural network with the dual ...
This paper presents a new efficient black-box attribution method based o...
Today's most advanced machine-learning models are hardly scrutable. The ...
A variety of methods have been proposed to try to explain how deep neura...
A multitude of explainability methods and theoretical evaluation scores ...
We describe a novel attribution method which is grounded in Sensitivity
...
The adoption of machine learning in critical contexts requires a reliabl...