Skip to content. | Skip to navigation

Sections
Personal tools
You are here: Home Theses A Blockchain Enabled Fast Consensus Mechanism in Federal Learning

Contact

Prof. Dr. S. Decker
RWTH Aachen
Informatik 5
Ahornstr. 55
D-52056 Aachen
Tel +49/241/8021501
Fax +49/241/8022321

How to find us

Annual Reports

Disclaimer

Webmaster

 

 

A Blockchain Enabled Fast Consensus Mechanism in Federal Learning

Thesis type
  • Master
Status Running
Supervisor(s)
Advisor(s)

Federated learning is a distributed machine learning paradigm aiming to collaboratively train a global model based on datasets across multiple clients and avoid the data privacy issues. With more clients and devices attending the federated learning network and the system heterogeneity including the storage, computational, and communication capabilities, under the traditional federated learning (server-client) network topology, the uncertainty of the client status raises the challenge of reaching a consensus global optimized model.


In recent years, blockchain is growing rapidly and applied in many fields, such as finance, Internet of Things and supply chain. In particular, the smart contract is a powerful component of blockchain that allows execution of arbitrary code on top of the blockchain, essentially a massively replicated eventually-consistent state machine. Considering the large number and various status of clients, smart contract coopeartes the output of each client, and it would be a carrier of federated learning consensus mechanism to achieve the goal. With the help of blockchain and the smart contract, there would be a promising solution to solve the consensus problem of federated learning, meanwhile keep the security and privacy.

 

This thesis focuses on the development of a blockchain-based efficient federated learning network consensus mechanism.

 

If you are interested in this thesis, a related topic or have additional questions, please do not hesitate to send a message to mou@dbis.rwth-aachen.de

Prerequisites

Knowledge of various programming languages, such as C++, Java, and Python will be fundamental.
Experience with cryptography and specific blockchain protocols, such as Hyperledger and Ethereum, will be very helpful.
Good communication skills.

Document Actions