Andres Felipe Posada Moreno

Andres Felipe Posada Moreno

Scientific Researcher and Doctoral student in eXplainable Artificial Intelligence

Institute for Data Science in Mechanical Engineering (DSME), RWTH Aachen University

Biography

Hi there! Since April 2021, I am a PhD student at the Institute for Data Science in Mechanical Engineering (DSME) at RWTH Aachen University under the supervision of Sebastian Trimpe, and an associate doctoral researcher at the Cluster of Excellence ‘Internet of Production’ funded by the DFG. Currently I’m waiting for my dissertation defense before continuing as a postdoc at the DSME until mid 2025.

Prior to that, I completed a bachelor’s degree in mechatronics engineering at the EIA University, a master’s degree in engineering at the Arts et Métiers ParisTech, worked on agroindustrial applications of IOT and AI at Asimetrix, and worked as a researcher on applied AI at the IMA Cybernetics Lab, RWTH Aachen University.

My research interests lie in the fields of explainable artificial intelligence (XAI) and applied AI. I am interested in the internal representations learned by neural networks and how these representations relate to human-understandable concepts. More specifically, I’m interested in methods for concept extraction and active learning and their usage in industrial applications. Recently I’ve started to explore foundation models, their latent representations and how they apply in industrial timeseries scenarios.

Interests
  • Artificial Intelligence
  • eXplainable Artificial Intelligence
  • Neural Networks
  • Computer Vision
Education

  • PhD in Artificial Intelligence, 2024

    RWTH Aachen University, Germany

  • M.Sc in General Engineering, Mechatronics Expertise, 2014

    Arts et Métiers ParisTech - École Nationale Supérieure d'Arts et Métiers, France

  • B.Sc in Mechatronics Engineering, 2012

    EIA University, Colombia

Experience

 
 
 
 
 
DSME - RWTH Aachen University
Scientific Researcher
DSME - RWTH Aachen University
April 2021 – Present Germany

Responsibilities include:

  • Lecturer: Lecturer on the subject “Learning-based Control Seminar”.
  • Student Supervision: Supervision of bachelor and master student thesis.
  • eXplainable Artificial Intelligence Research: Specialized in global explainability methods (concept extraction) for deep neural networks.
  • Organization of Seminar series: Organization of the “Research Seminar on AI” RSAI.
  • Excellence cluster: Co-lead of the AI expert group at the excellence cluster “Internet of Production”.
 
 
 
 
 
IMA - RWTH Aachen University
Scientific Researcher
IMA - RWTH Aachen University
April 2018 – April 2021 Germany

Responsibilities include:

  • Lecturer: Lecturer on the subject “Artificial Intelligence and Data Analytics for Engineers”.
  • Student Supervision: Supervision of bachelor and master student thesis.
  • Applied Artificial Intelligence Research: Specialized in the application of neural networks to high-dimensional data scenarios.
 
 
 
 
 
Asimetrix
Technology Analyst
Asimetrix
October 2014 – October 2017 Colombia

Responsibilities include:

  • Research and Development: Focused on AI and IoT solutions in the protein production industry. Specialized in the development and testing of both IoT devices and cloud based AI services.
  • Conference Speaker: Delivered presentations on emerging technologies and data-driven decision-making processes.
 
 
 
 
 
EIA University
Part-time lecturer
EIA University
July 2014 – November 2014 Colombia

Responsibilities include:

  • Teaching: Lecturer on the “Mechatronics Design” lecture for 8th semester undergraduate students.

Projects

*
Explaining Neural Networks Through Concept Extraction
This research project focuses on the methodological and application aspects of how to explain what a deep neural network learns through training. Specifically, this project tackles the development of methods for concept extraction, localization, and learning, as well as their applications in industrial scenarios.
Explaining Neural Networks Through Concept Extraction
Internet of Production (IoP)
The vision of the IoP was to enable a new level of cross-domain collaboration by providing context-aware data and AI methods from production, development and usage in real-time on an appropriate level of granularity.
DFG RWTH
Internet of Production (IoP)
Increasing the quality of rail transport through data-based damage pattern recognition on rail vehicles - QUISS
The aim of the QUISS project was to develop data-based applications using modern data science and artificial intelligence approaches to optimize or support the scheduling and maintenance of rail vehicles.
BMDV
Increasing the quality of rail transport through data-based damage pattern recognition on rail vehicles - QUISS

Recent Publications

Quickly discover relevant content by filtering publications.
Pointwise-in-Time Diagnostics for Reinforcement Learning during Training and Runtime
A framework for explaining RL agents during training and runtime based the definition of an agents desired behavior through linear temporal logic (LTL).
Noel Brindise, Andres Felipe Posada-Moreno, Cedric Langbort, Sebastian Trimpe
Cite URL
ECLAD: Extracting Concepts with Local Aggregated Descriptors
A novel method for automatic concept extraction and localization based on representations obtained through pixel-wise aggregations of CNN activation maps.
Andres Felipe Posada-Moreno, Nikita Surya, Sebastian Trimpe
Cite Project Project DOI URL
Scale-Preserving Automatic Concept Extraction (SPACE)
We introduce the Scale-Preserving Automatic Concept Extraction (SPACE) algorithm, as a state-of-the-art alternative concept extraction technique for CNNs, focused on industrial applications.
Andres Felipe Posada-Moreno, Lukas Kreisköther, Tassilo Glander, Sebastian Trimpe
Cite Project Project DOI URL
Actionable Artificial Intelligence for the Future of Production
Compilation of fundamentals and applications of AI and data based systems in the context of the Industry 4.0.
Mohamed Behery, Philipp Brauner, Hans Aoyang Zhou, Merih Seran Uysal, Vladimir Samsonov, Martin Bellgardt, Florian Brillowski, Tobias Brockhoff, Anahita Farhang Ghahfarokhi, Lars Gleim, Leon Michel Gorissen, Marco Grochowski, Thomas Henn, Elisa Iacomini, Thomas Kaster, István Koren, Martin Liebenberg, Leon Reinsch, Liam Tirpitz, Minh Trinh, Andres Felipe Posada-Moreno, Luca Liehner, Thomas Schemmer, Luisa Vervier, Marcus Völker, Philipp Walderich, Song Zhang, Christian Brecher, Robert H. Schmitt, Stefan Decker, Thomas Gries, Constantin Leon Häfner, Michael Herty, Matthias Jarke, Stefan Kowalewski, Torsten W. Kuhlen, Johannes Henrich Schleifenbaum, Sebastian Trimpe, Wil M. P. van der Aalst, Martina Ziefle, Gerhard Lakemeyer
Cite Project DOI URL
Objectifying Machine Setup and Parameter Selection in Expert Knowledge Dependent Industries Using Invertible Neural Networks
Using invertible neural networks and expert knowledge for parameter prediction in industrial applications.
Kai Müller, Andrés Posada-Moreno, Lukas Pelzer, Thomas Gries
Cite Project DOI
See all publications

Contact