Neuro-symbolic AI combines the strengths of neural and symbolic methods to efficiently learn and reason over models of the world. Typically, many of the assurances that can be provided by such systems are limited to the learning errors due to the neural component. In this Ph.D. project, you will be exploring the use of Physics-Informed Neural Networks to encode the symbolic knowledge into the learned neural models of the world to create a Neuro-symbolic AI that can provide assurance guarantees. This project will be a part of the "Hierarchical Abstractions And Reasoning for Neuro Symbolic System" project team.
Required knowledge
A successful candidate should have:
- a degree in Computer Science, Information Technology Engineering, or equivalent,
- excellent mathematical and analytical skills,
- excellent skills in machine learning (i.e., deep learning, PINNs) and AI,
- excellent communication skills (i.e., both written and verbal),
- the ability to work independently, and
- the ability to collaborate with a team of researchers and engineers.
Project funding
Other
