Master Thesis Causal Discovery for Vehicle Quality Data using Transformers
Master Thesis Causal Discovery for Vehicle Quality Data using Transformer Models
In recent years, the development of attention-based deep learning models has significantly enhanced our ability to interpret data and draw causal graphs, explaining the relationships between variables. These models surpass traditional causal discovery algorithms in terms of computational efficiency and the precision of inferred graph structures, particularly in high-dimensional datasets. This thesis will focus on using Transformer models to understand why certain error codes and environmental conditions in field vehicles might lead to critical failures.
What awaits you?
Engage in state-of-the-art research on Transformer models for causal discovery.
Work with real-world data sets to draw applicative conclusions.
Push the boundaries of predictive maintenance using deep learning approaches.
Gain practical AI experience by implementing PyTorch and Hugging face models.
Utilize our cutting-edge training infrastructure and platforms to conduct experiments and evaluate your approach efficiently.
Present the thesis orally and through a research paper.
Please note that you must ensure that the thesis is supervised by a university.
What should you bring along?
- You are a master student approaching the end of your degree in computer science or related fields with focus on machine learning or artificial intelligence.
- You have strong knowledge in deep learning, transformers, graphs and NLP.
- You have excellent programming skills in Python, PyTorch and worked with modern cloud environment tools such as AWS.
- You speak English fluently.
- You are driven by curiosity and motivated to solve scientific problems independently as well as sharing ideas and working in a team.
