SHK: Distributed Machine Learning
Distributed Machine Learning
Centralized computation for machine learning (ML) tasks can be compute intensive and also requires input data that might not be available for sharing due to various reasons such as privacy or available communication bandwidth. This gap is filled by the concept of distributed machine learning that refers to multi-node machine learning systems that bring the model to the data (not the other way around) and hence train, collect and aggregate model instances in repetitive synchronization steps.
For more information on Flower : A Friendly Federated Learning Framework
- Completed or curretly doing a bachelor's degree in computer science or electrical engineering and information technology.
- Hands-on expirence with Linux
- Programming Knowledge in Python
- Good English writing skills
- Implementation of Federated Machine Learning (FML) in an automotive environment
Using Flower (FLM framework) to execute and evaluate FML tasks
Design and code APIs for node interactions in Flower.
As soon as possible
Up to 1 year
10 hours per week