SAMBA: A Generic Framework for Secure Federated Multi-Armed Bandits (Extended Abstract)

We tackle the problem of secure cumulative reward maximization in multi-armed bandits in a cross-silo federated learning setting. Under the orchestration of a central server, each data owner participating at the cumulative reward computation has the guarantee that its raw data is not seen by some other participant. We rely on cryptographic schemes and propose SAMBA, a generic framework for Secure federAted Multi-armed BAndits. We show that SAMBA returns the same cumulative reward as the non-secure versions of bandit algorithms, while satisfying formally proven security properties. We also show that the overhead due to cryptographic primitives is linear in the size of the input, which is confirmed by our implementation.