Master Thesis: Knowledge Graphs for Structured and Unstructured Data
There are several topics available on Knowledge Graph embedding methods for structured or unstructured data such as images (intersection with Computer Vision: Scene Graph Generation). The candidate is expected to satisfy the following requirements:
- Theoretical understanding and hands-on experience with deep learning methods.Preferably having good grades from Deep Learning or Machine Learning course at LMU.
- Familiar with Tensorflow, Pytorch or Keras.
- Good understanding of Linear Algebra and Statistics. Specifically I recommend you to recap your linear algebra knowledge here (this series has an overall length of 2 and a half hours).
- I would highly recommend you to go through this paper before meeting.
Contact: Sahand Sharifzadeh (sharifzadeh@dbs.ifi.lmu.de)