Search Results for author:
Found 3 papers, 0 papers with code
FedPerm: Private and Robust Federated Learning by Parameter Permutation
We present FedPerm, a new FL algorithm that addresses both these problems by combining a novel intra-model parameter shuffling technique that amplifies data privacy, with Private Information Retrieval (PIR) based techniques that permit cryptographic aggregation of clients' model updates.
Optimizing Inference Performance of Transformers on CPUs
The Transformer architecture revolutionized the field of natural language processing (NLP).
Hemlock : Compact and Scalable Mutual Exclusion
We present Hemlock, a novel mutual exclusion locking algorithm that is extremely compact, requiring just one word per thread plus one word per lock, but which still provides local spinning in most circumstances, high throughput under contention, and low latency in the uncontended case.
Distributed, Parallel, and Cluster Computing Operating Systems D.4.1
