tf_privacy.v1.DPAdamGaussianOptimizer

DP subclass of tf.compat.v1.train.AdamOptimizer.

tf_privacy.v1.DPAdamGaussianOptimizer(
    l2_norm_clip,
    noise_multiplier,
    num_microbatches=None,
    unroll_microbatches=False,
    *args,
    **kwargs
)

You can use this as a differentially private replacement for tf.compat.v1.train.AdamOptimizer. This optimizer implements DP-SGD using the standard Gaussian mechanism.

When instantiating this optimizer, you need to supply several DP-related arguments followed by the standard arguments for AdamOptimizer.

Examples:

# Create optimizer.
opt = DPAdamGaussianOptimizer(l2_norm_clip=1.0, noise_multiplier=0.5, num_microbatches=1,
         <standard arguments>)

When using the optimizer, be sure to pass in the loss as a rank-one tensor with one entry for each example.

# Compute loss as a tensor. Do not call tf.reduce_mean as you
# would with a standard optimizer.
loss = tf.nn.sparse_softmax_cross_entropy_with_logits(
    labels=labels, logits=logits)

train_op = opt.minimize(loss, global_step=global_step)

Args
l2_norm_clip	Clipping norm (max L2 norm of per microbatch gradients).
noise_multiplier	Ratio of the standard deviation to the clipping norm.
num_microbatches	Number of microbatches into which each minibatch is split. If None, will default to the size of the minibatch, and per-example gradients will be computed.
unroll_microbatches	If true, processes microbatches within a Python loop instead of a tf.while_loop. Can be used if using a tf.while_loop raises an exception.
*args	These will be passed on to the base class __init__ method.
**kwargs	These will be passed on to the base class __init__ method.

Methods

get_config

View source

get_config()

Creates configuration for Keras serialization.

This method will be called when Keras creates model checkpoints and is necessary so that deserialization can be performed.

Returns
A dict object storing arguments to be passed to the init method upon deserialization.

Class Variables
GATE_GRAPH	2
GATE_NONE	0
GATE_OP	1