Introduces formal verification to compute certified neuron range bounds for CKKS-encrypted neural networks, eliminating overflow failures that previously reached 47%.
On the number of nonscalar multiplications necessary to evaluate polynomials.SIAM Journal on Computing, 2(1):60–66
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Encrypted Neural Networks without Overflows
Introduces formal verification to compute certified neuron range bounds for CKKS-encrypted neural networks, eliminating overflow failures that previously reached 47%.