Privacy-Preserving Federated Learning through Clustered Sampling on Fine-Tuning Distributed non-iid Large Language Models.
Sheng YunMd. Zakirul Alam BhuiyanMd. Taufiq Al Hasib SadiShen SuPublished in: ISPA/BDCloud/SocialCom/SustainCom (2023)
Keyphrases
- privacy preserving
- language model
- fine tuning
- multi party
- horizontally partitioned data
- probabilistic model
- language modeling
- partitioned data
- n gram
- digital libraries
- statistical language models
- language modelling
- privacy sensitive
- bayesian networks
- sensitive data
- privacy preserving data mining
- data privacy
- privacy preservation
- distributed data
- private information
- query expansion
- vertically partitioned data
- secure multiparty computation
- active learning
- horizontally partitioned