| 1. |
- Chen, Jialu, et al.
(författare)
-
Lightweight Privacy-preserving Training and Evaluation for Discretized Neural Networks
- 2020
-
Ingår i: IEEE Internet of Things Journal. - : IEEE. - 2327-4662. ; 7:4, s. 2663-2678
-
Tidskriftsartikel (refereegranskat)abstract
- Machine learning, particularly the neural network, is extensively exploited in dizzying applications. In order to reduce the burden of computing for resource-constrained clients, a large number of historical private datasets are required to be outsourced to the semi-trusted or malicious cloud for model training and evaluation. To achieve privacy preservation, most of the existing work either exploited the technique of public key fully homomorphic encryption (FHE) resulting in considerable computational cost and ciphertext expansion, or secure multiparty computation (SMC) requiring multiple rounds of interactions between user and cloud. To address these issues, in this paper, a lightweight privacy-preserving model training and evaluation scheme LPTE for discretized neural networks is proposed. Firstly, we put forward an efficient single key fully homomorphic data encapsulation mechanism (SFH-DEM) without exploiting public key FHE. Based on SFH-DEM, a series of atomic calculations over the encrypted domain including multivariate polynomial, nonlinear activation function, gradient function and maximum operations are devised as building blocks. Furthermore, a lightweight privacy-preserving model training and evaluation scheme LPTE for discretized neural networks is proposed, which can also be extended to convolutional neural network. Finally, we give the formal security proofs for dataset privacy, model training privacy and model evaluation privacy under the semi-honest environment and implement the experiment on real dataset MNIST for recognizing handwritten numbers in discretized neural network to demonstrate the high efficiency and accuracy of our proposed LPTE.
|
|
| 2. |
- Wang, Yucheng, et al.
(författare)
-
Late Quaternary Dynamics of Arctic Biota from Ancient Environmental Genomics
- 2021
-
Ingår i: Nature. - : Springer Science and Business Media LLC. - 0028-0836 .- 1476-4687. ; 600:7887, s. 86-92
-
Tidskriftsartikel (refereegranskat)abstract
- During the last glacial–interglacial cycle, Arctic biotas experienced substantial climatic changes, yet the nature, extent and rate of their responses are not fully understood1–8. Here we report a large-scale environmental DNA metagenomic study of ancient plant and mammal communities, analysing 535 permafrost and lake sediment samples from across the Arctic spanning the past 50,000 years. Furthermore, we present 1,541 contemporary plant genome assemblies that were generated as reference sequences. Our study provides several insights into the long-term dynamics of the Arctic biota at the circumpolar and regional scales. Our key fndings include: (1) a relatively homogeneous steppe–tundra fora dominated the Arctic during the Last Glacial Maximum, followed by regional divergence of vegetation during the Holocene epoch; (2) certain grazing animals consistently co-occurred in space and time; (3) humans appear to have been a minor factor in driving animal distributions; (4) higher efective precipitation, as well as an increase in the proportion of wetland plants, show negative efects on animal diversity; (5) the persistence of the steppe–tundra vegetation in northern Siberia enabled the late survival of several now-extinct megafauna species, including the woolly mammoth until 3.9 ± 0.2 thousand years ago (ka) and the woolly rhinoceros until 9.8 ± 0.2 ka; and (6) phylogenetic analysis of mammoth environmental DNA reveals a previously unsampled mitochondrial lineage. Our fndings highlight the power of ancient environmental metagenomics analyses to advance understanding of population histories and long-term ecological dynamics
|
|
| 3. |
- Wang, Yucheng, et al.
(författare)
-
Reply to: When did mammoths go extinct?
- 2022
-
Ingår i: Nature. - : Springer Science and Business Media LLC. - 0028-0836 .- 1476-4687. ; 612:7938, s. 4-6
-
Tidskriftsartikel (refereegranskat)
|
|
