Privacy-Preserving Stream Aggregation with Fault Tolerance

By T-H. Hubert Chan, Elaine Shi, Dawn Song. 2012

229 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Privacy & Security

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Privacy & Security

Citations in the corpus

Advances and Open Problems in Federated Learning (2019, background): “There is a rich literature exploring secure aggregation in both the single-server setting (via pairwise additive masking [12, 188, 73], via threshold homomorphic encryption [356, 193, 92], and via generic secure multi-party computation [86]) as well as in the multiple non-colluding servers setting [”
Privacy Amplification by Decentralization (2020, background): “Unfortunately, LDP comes at a great cost in utility: for real summation with n users, the best possible error under LDP is a factor √ n larger than in the centralized (trusted curator) model of DP (Chan et al., 2012a).”
Distributed Differential Privacy via Shuffling (2018, neutral): “[9], Bonawitz et al.”
A Hybrid Approach to Privacy-Preserving Federated Learning (2018, background): “Approaches for the private collection of streaming data, including [2, 8, 21, 37], aim to recover computation when one or more parties go down.”
Efficient Private Statistics with Succinct Sketches (2015, background): “Chan et al. [15] also provides fault tolerance by extending [61]’s protocol, however, with a poly-logarithmic penalty.”
A Lightweight Privacy-Preserving Data Aggregation Scheme for Fog Computing-Enhanced IoT (2017, background): “[17] presented novel mechanisms to enhance Shi et al.”
Lightweight Crypto-Assisted Distributed Differential Privacy for Privacy-Preserving Distributed Learning (2020, background): “Notice that all the existing fault tolerance schemes come with a trade-off [21], [18], [28].”
A Scalable Scheme for Privacy-Preserving Aggregation of Time-Series Data (2013, background): “An important challenge left open in [25] (and also discussed in [8]) is that of efficiently computing sums over large plaintext spaces.”
Computational Differential Privacy from Lattice-Based Cryptography (2017, background): “In a subsequent work by Chan et al. [10], this connection is still not completely established, since the polynomial-time reduction between an attacker against a secure PSA scheme and a database distinguisher is missing.”
Encrypted data processing with Homomorphic Re-Encryption (2017, background): “[11, 26] also studied encrypted data aggregation in the presence of a distrusted aggregator.”
Practical Secure Aggregation for Privacy-Preserving Machine Learning (2017, methodology): “[14] use an approach similar to ours, but in the exponent in some group”
Achieving Differential Privacy in Secure Multiparty Data Aggregation Protocols on Star Networks (2017, background): “Specifically, although the prior art could also be extended to compute linear combinations, in most cases ([1, 3, 4, 9, 12, 15, 16, 22]), the protocols have to reveal the weights ci to the participants.”
Differentially private data aggregation with optimal utility (2014, background): “Many schemes [5, 14, 38, 39] involve a signiﬁ-cant computational eﬀort on the user’s side, making them impractical in several scenarios, e.g., for aggregating data stored on mobile devices with limited computation power.”
Separating Computational and Statistical Differential Privacy in the Client-Server Model (2016, unknown): “A number of works [CSS12, GMPS13, HOZ13, KMS14, GKM + 16] have since sought to characterize the extent to which CDP yields accuracy improvements for two-party privacy problems.”
Efficient and Privacy-Enhanced Federated Learning for Industrial Artificial Intelligence (2020, methodology): “Besides, Chan et al. [40] and Bonawitz et al. [20] address the problem of dropouts, e.g., completing the protocol even if a user drops out.”
Fault-Tolerant Privacy-Preserving Statistics (2012, methodology): “In [2] data producers also communicate unidirectionally with data consumers but have higher communication cost.”
Cryptϵ: Crypto-Assisted Differential Privacy on Untrusted Servers (2019, methodology): “with the trust assumption of the CDP model is using cryptographic primitives [9, 14, 17, 20, 29, 31, 37, 41, 92, 93].”