Blogs on deep neural networks » Lamarr Institute

00 Blog hanxiao tan punktewolken - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Explainability Methods for Point Clouds II

14. June 2023

Two different approaches to point cloud interpretability are proposed, which elucidate the properties of the model from global and model-internal perspectives, respectively. All toolkits are now available on GitHub....

00 Blog Paass Foundation Modelle als Grundlage von KI Systemen - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Research

Creating coherent texts with language models: Foundation Models as the basis for AI systems

10. August 2022

So-called foundation models are developing rapidly and can already perform qualitatively demanding tasks automatically and view different media simultaneously. This article takes a look behind the scenes of the major language models....

00 Blog Sujan Semantic Concept Testing 9 - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Application

An Approach to Granular Testing of Deep Neural Networks

01. June 2022

Establishing rigorous testing and certification techniques is essential before deploying new technologies like Deep Learning (DL) in safety-critical applications. We propose a testing approach that could identify weaknesses in DL models....

00 Blog Tan Punktewolke 9 - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Research

An Explainability Method for Point Clouds

04. May 2022

A surrogate model-based explainability approach is proposed for point cloud classifiers together with two different evaluation metrics validating the plausibility of the approach. The toolkit is now available online....

00 Blog Paass DL - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Basics

Deep Learning: How do deep neural networks work?

21. April 2021

Deep Learning uses deep neural networks to recognize images, understand text and make decisions more accurately. To analyze large data sets, "layered learning" uses multiple hidden layers in the network....