secretflow.security.privacy.accounting#

secretflow.security.privacy.accounting.gdp_accountant#

Functions:

`cal_mu_uniform`(step, noise_multiplier, ...)	Calculate mu from uniform subsampling.
`cal_mu_poisson`(step, noise_multiplier, ...)	Calculate mu from Poisson subsampling.
`delta_gaussian`(eps, mu)	Calculate dual between mu-GDP and (epsilon, delta)-DP.
`get_eps_from_mu`(mu, delta)	Get epsilon from mu given delta via inverse dual.

secretflow.security.privacy.accounting.gdp_accountant.cal_mu_uniform(step: int, noise_multiplier, num_samples, batch_size)[源代码]#

Calculate mu from uniform subsampling.

参数:

step – The current step of model training or prediction.
epnoise_multipliers – Epsilon for pure DP.
batch_size – Batch size.
num_samples – Number of all samples.

secretflow.security.privacy.accounting.gdp_accountant.cal_mu_poisson(step: int, noise_multiplier, num_samples, batch_size)[源代码]#

Calculate mu from Poisson subsampling.

参数:

step – The current step of model training or prediction.
epnoise_multipliers – Epsilon for pure DP.
batch_size – Batch size.
num_samples – Number of all samples.

secretflow.security.privacy.accounting.gdp_accountant.delta_gaussian(eps, mu)[源代码]#

Calculate dual between mu-GDP and (epsilon, delta)-DP.

参数:

mu – The parameters of the GDP.
eps – The parameters of the (epsilon, delta)-DP.

secretflow.security.privacy.accounting.gdp_accountant.get_eps_from_mu(mu, delta)[源代码]#

Get epsilon from mu given delta via inverse dual.

参数:

mu – The parameters of the GDP.
delta – The parameters of the (epsilon, delta)-DP.

secretflow.security.privacy.accounting.log_utils#

Functions:

`add_log`(logx, logy)	Add two numbers in the log space.
`sub_log`(logx, logy)	Subtract two numbers in the log space.
`erfc_log`(x)	Calculate log(erfc(x)) with high accuracy for large x.
`comb_log`(n, k)
`log_alpha_int`(q, sigma, alpha)	Calculate log(A_alpha) for integer alpha.
`log_alpha_frac`(q, sigma, alpha)	Calculate log(A_alpha) for fractional alpha.
`log_alpha`(q, sigma, alpha)	Calculate log(A_alpha) for any positive finite alpha.

secretflow.security.privacy.accounting.log_utils.add_log(logx, logy)[源代码]#: Add two numbers in the log space.

secretflow.security.privacy.accounting.log_utils.sub_log(logx, logy)[源代码]#: Subtract two numbers in the log space. The return value must be non-negative.

secretflow.security.privacy.accounting.log_utils.erfc_log(x)[源代码]#: Calculate log(erfc(x)) with high accuracy for large x.

secretflow.security.privacy.accounting.log_utils.comb_log(n, k)[源代码]#

secretflow.security.privacy.accounting.log_utils.log_alpha_int(q, sigma, alpha)[源代码]#: Calculate log(A_alpha) for integer alpha. 0 < q < 1.

secretflow.security.privacy.accounting.log_utils.log_alpha_frac(q, sigma, alpha)[源代码]#: Calculate log(A_alpha) for fractional alpha. 0 < q < 1.

secretflow.security.privacy.accounting.log_utils.log_alpha(q, sigma, alpha)[源代码]#: Calculate log(A_alpha) for any positive finite alpha.

secretflow.security.privacy.accounting.rdp_accountant#

Functions:

`rdp_core`(q, noise_multiplier, alpha)	Comnpute RDP of the Sampled Gaussian mechanism at order alpha.
`get_rdp`(q, noise_multiplier, steps, orders)	Calculate RDP of the Sampled Gaussian Mechanism.
`cal_delta`(orders, rdp, eps)	Calculate delta given a list of RDP values and target epsilon.
`cal_eps`(orders, rdp, delta)	Calculate epsilon given a list of RDP values and target delta.
`get_privacy_spent_rdp`(orders, rdp[, ...])	Calculates delta (or eps) for given eps (or delta) from RDP values.

secretflow.security.privacy.accounting.rdp_accountant.rdp_core(q: float, noise_multiplier: float, alpha: float)[源代码]#

Comnpute RDP of the Sampled Gaussian mechanism at order alpha.

参数:

q – The sampling rate.
noise_multiplier – The noise_multiplier used for calculating the std of the additive Gaussian noise.
alpha – The order at which RDP is cald

返回:

RDP at alpha, can be np.inf.

secretflow.security.privacy.accounting.rdp_accountant.get_rdp(q: float, noise_multiplier: float, steps: int, orders)[源代码]#

Calculate RDP of the Sampled Gaussian Mechanism.

参数:

q – The sampling rate.
noise_multiplier – The ratio of the standard deviation of the Gaussian noise to the l2-sensitivity of the function to which it is added
steps – The number of steps.
orders – An array (or a scalar) of RDP orders.

返回:

The RDPs at all orders. Can be np.inf.

secretflow.security.privacy.accounting.rdp_accountant.cal_delta(orders, rdp, eps: float)[源代码]#

Calculate delta given a list of RDP values and target epsilon. :param orders: An array (or a scalar) of orders. :param rdp: A list (or a scalar) of RDP guarantees. :param eps: The target epsilon.

返回:: Pair of (delta, optimal_order).
抛出:: ValueError – If input is malformed.

secretflow.security.privacy.accounting.rdp_accountant.cal_eps(orders, rdp, delta: float)[源代码]#

Calculate epsilon given a list of RDP values and target delta. :param orders: An array (or a scalar) of orders. :param rdp: A list (or a scalar) of RDP guarantees. :param delta: The target delta.

返回:: Pair of (eps, optimal_order).
抛出:: ValueError – If input is malformed.

secretflow.security.privacy.accounting.rdp_accountant.get_privacy_spent_rdp(orders, rdp, target_eps: Optional[float] = None, target_delta: Optional[float] = None)[源代码]#

Calculates delta (or eps) for given eps (or delta) from RDP values.

参数:

orders – An array (or a scalar) of RDP orders.
rdp – An array of RDP values. Must be of the same length as the orders list.
target_eps – If not None, the epsilon for which we cal the corresponding delta.
target_delta – If not None, the delta for which we cal the corresponding epsilon. Exactly one of target_eps and target_delta must be None.

返回:

A tuple of epsilon, delta, and the optimal order.

抛出:

ValueError – If target_eps and target_delta are messed up.