Master Thesis Exploring Formal Guarantees for Neural Network Optimizations

The efficient deployment of complex neural networks on modern edge and embedded hardware presents a key challenge. Here, unique opportunities await you to overcome this bottleneck and actively shape the future of AI deployment. Are you ready to make a decisive contribution with your thesis and redefine the boundaries of what is technically feasible? • During your Thesis, you delve deeply into current research, analyzing state-of-the-art optimization strategies for the deployment of neural networks to assess their suitability for efficient HW/SW co-design. • Furthermore, you develop an innovative mathematical framework to formally verify numerical tolerance limits for neural network operations, thereby establishing a new verification metric. • You evaluate the complex interplay between hardware acceleration and numerical stability to identify and integrate optimal trade-offs. • Lastly, you document your research findings with precision and present them persuasively to the team, actively fostering technical exchange. • Education: Studies in the field of Electrical Engineering, Computer Science, or comparable • Experience and Knowledge: strong experience in Python/C++, neural network compiler (TVM), basic hardware knowledge, Deep Learning Concepts, Linux, LaTeX • Personality and Working Practice: you possess a strong inventive spirit and have the desire to discuss and communicate complex research topics in a global environment. • Work Routine: office attendance required • Languages: good in English Start: according to prior agreement Duration: 6 months Requirement for this thesis is the enrollment at university. Please attach your CV, transcript of records, examination regulations and if indicated a valid work and residence permit. Diversity and inclusion are not just trends for us but are firmly anchored in our corporate culture. Therefore, we welcome all applications, regardless of gender, age, disability, religion, ethnic origin or sexual identity. Need further information about the job? Shubham Rai (Functional Department)  Work #LikeABosch starts here: Apply now! #LI-DNI