Search Results for author: Helge Spieker

Found 17 papers, 5 papers with code

Probabilistic Model Checking of Stochastic Reinforcement Learning Policies

no code implementations27 Mar 2024 Dennis Gross, Helge Spieker

We introduce a method to verify stochastic reinforcement learning (RL) policies.

Enhancing Manufacturing Quality Prediction Models through the Integration of Explainability Methods

no code implementations27 Mar 2024 Dennis Gross, Helge Spieker, Arnaud Gotlieb, Ricardo Knoblauch

This research presents a method that utilizes explainability techniques to amplify the performance of machine learning (ML) models in forecasting the quality of milling processes, as demonstrated in this paper through a manufacturing use case.

Towards Trustworthy Automated Driving through Qualitative Scene Understanding and Explanations

no code implementations25 Mar 2024 Nassim Belmecheri, Arnaud Gotlieb, Nadjib Lazaar, Helge Spieker

In this article, we introduce the Qualitative Explainable Graph (QXG), which is a unified symbolic and qualitative representation for scene understanding in urban mobility.

Scene Understanding

Detecting Intentional AIS Shutdown in Open Sea Maritime Surveillance Using Self-Supervised Deep Learning

no code implementations24 Oct 2023 Pierre Bernabé, Arnaud Gotlieb, Bruno Legeard, Dusica Marijan, Frank Olaf Sem-Jacobsen, Helge Spieker

In maritime traffic surveillance, detecting illegal activities, such as illegal fishing or transshipment of illicit products is a crucial task of the coastal administration.

Acquiring Qualitative Explainable Graphs for Automated Driving Scene Interpretation

1 code implementation24 Aug 2023 Nassim Belmecheri, Arnaud Gotlieb, Nadjib Lazaar, Helge Spieker

The future of automated driving (AD) is rooted in the development of robust, fair and explainable artificial intelligence methods.

Explainable artificial intelligence

Testing Deep Learning Models: A First Comparative Study of Multiple Testing Techniques

no code implementations24 Feb 2022 Mohit Kumar Ahuja, Arnaud Gotlieb, Helge Spieker

To discover faults in DL models, existing software testing methods have been adapted and refined accordingly.

Autonomous Driving

Predictive Machine Learning of Objective Boundaries for Solving COPs

no code implementations4 Nov 2021 Helge Spieker, Arnaud Gotlieb

Solving Constraint Optimization Problems (COPs) can be dramatically simplified by boundary estimation, that is, providing tight boundaries of cost functions.

BIG-bench Machine Learning

Mistake-driven Image Classification with FastGAN and SpinalNet

no code implementations29 Sep 2021 Mohit Kumar Ahuja, Sahil Sahil, Helge Spieker

Image classification with classes of varying difficulty can cause performance disparity in deep learning models and reduce the overall performance and reliability of the predictions.

Classification Data Augmentation +1

Constraint-Guided Reinforcement Learning: Augmenting the Agent-Environment-Interaction

1 code implementation24 Apr 2021 Helge Spieker

Still, training the agents is data-intensive and there are no guarantees that the learned behavior is safe and does not violate rules of the environment, which has limitations for the practical deployment in real-world scenarios.

reinforcement-learning Reinforcement Learning (RL)

Learning Objective Boundaries for Constraint Optimization Problems

no code implementations20 Jun 2020 Helge Spieker, Arnaud Gotlieb

Constraint Optimization Problems (COP) are often considered without sufficient knowledge on the boundaries of the objective variable to optimize.

Time-aware Test Case Execution Scheduling for Cyber-Physical Systems

no code implementations12 Feb 2019 Morten Mossige, Arnaud Gotlieb, Helge Spieker, Hein Meling, Mats Carlsson

When testing industrial robots, it is common that the target machines need to share some common resources, e. g., costly hardware devices, and so there is a need to schedule test case execution on the target machines, accounting for these shared resources.

Industrial Robots Scheduling

Towards Testing of Deep Learning Systems with Training Set Reduction

no code implementations14 Jan 2019 Helge Spieker, Arnaud Gotlieb

However, testing the training routines requires running them and fully training a deep learning model can be resource-intensive, when using the full data set.

ITE: A Lightweight Implementation of Stratified Reasoning for Constructive Logical Operators

1 code implementation9 Nov 2018 Arnaud Gotlieb, Dusica Marijan, Helge Spieker

Constraint Programming (CP) is a powerful declarative programming paradigm where inference and search are interleaved to find feasible and optimal solutions to various type of constraint systems.

Negation

Multi-Cycle Assignment Problems with Rotational Diversity

no code implementations8 Nov 2018 Helge Spieker, Arnaud Gotlieb, Morten Mossige

Multi-cycle assignment problems address scenarios where a series of general assignment problems has to be solved sequentially.

Cannot find the paper you are looking for? You can Submit a new open access paper.