Master Thesis Project - Performance evaluation and benchmarking of federated vs centralized learning i Lund

Company... At Bosch, we shape the future by inventing high-quality technologies and services that spark enthusiasm and enrich people’s lives. Our promise to our associates is rock-solid: we grow together, we enjoy our work, and we...

Om tjänsten

Company Description

At Bosch, we shape the future by inventing high-quality technologies and services that spark enthusiasm and enrich people’s lives. Our promise to our associates is rock-solid: we grow together, we enjoy our work, and we inspire each other. Join in and feel the difference.

Bosch R&D Center Lund stands for modern development in cutting edge technology in the areas of connectivity, security, mobility solutions and AI. We are growing rapidly and looking for people to join us on our mission to become the Bosch Group’s 1st address for secure connected mobility solutions. We are working on a range of interesting projects, with a particular focus on software development for the automotive industry, electrical bicycles and Internet of Things.

Job Description

Problem statement
Federated learning is a technique for optimizing deep learning networks in a decentralized setting. Using an inequal amount of data across many nodes, the goal is to train the central model to a high standard.

With the increased use of federated learning comes the necessity of validating performances of resulting models in comparison to locally trained models. Does the data split of federated learning come with a cost, when comparing “normal” metrics, and which workflow generalizes the data best? Does each regional model bias the concatenated model with their locally derived data?

Federated learning also has some extended metrics for evaluating its performance. Can we use it to somehow compare to a locally trained model?

Proposed solution:
1. The thesis project shall focus on understanding the benefits and drawbacks with federated learning.
2. Proposed steps could include:
3. Study and learn the theory and best practices within federated learning.
4. Set up a system, including master and child devices, to enable simulation of federated learning.
5. Pick at least one suitable open dataset that contains bias and other interesting characteristics for federated learning.
6. Conduct federated learning using the prototype system and try out different learning strategies.
7. Identify and apply useful metrics to evaluate and compare the learned models.
8. Document theory and findings from practical work in a thesis report.

Scope of master thesis project:
Two students completing 30 credits each (20 weeks) onsite at the Lund office

Qualifications

In order to be successful in the project with think you are:

  • A student in Information Technology, Computer Science, Electronics, Math or Physics at LTH.
  • Experienced with or have at least some knowledge of programming in Python or R, and Linux.
  • Experienced with deep learning and image processing.

Supervisors:
Bosch: Adam Koch at ESW12

Additional Information

Your future job location offers you:

Flexible work time options, benefits and services, medical services, employee discounts, various sports and health opportunities, catering facilities, collective agreement, wellness contribution, access to local public transport, and room for creativity.
Diversity is our strength! At Bosch we look at diversity in gender, generation, nationalities and culture as our advantage. We believe mixed teams to be more successful because they utilize the potential offered by different perspectives and solution strategies. We therefore promote mixed teams at all levels and draw on the entire talent pool.

Publicerad den

29-11-2022

Extra information

Status
Öppen
Ort
Lund
Körkort önskas
Nej
Tillgång till bil önskas
Nej
Personligt brev krävs
Nej
Ansök nu

Dela jobbannonsen