jax_privacy.accounting.truncated_dpsgd_event

jax_privacy.accounting.truncated_dpsgd_event(noise_multiplier, iterations, *, sampling_prob, num_examples, truncated_batch_size)[source]

Returns the DpEvent for truncated DP-SGD with the given training params.

This mechanism is like DP-SGD, but batches larger than truncated_batch_size are truncated to size truncated_batch_size. See these references for more information about this mechanism:

[Scalable DP-SGD: Shuffling vs. Poisson Subsampling](https://arxiv.org/abs/2411.04205)
[Tighter Privacy Analysis for Truncated Poisson Sampling](https://arxiv.org/abs/2508.15089)

Parameters:

noise_multiplier (float) – The noise multiplier of the mechanism.
iterations (int) – The number of iterations to run the mechanism for.
sampling_prob (float) – The Poisson sampling probability of the mechanism, i.e., the probability an example will be included in each batch before truncation.
num_examples (int) – The number of examples in the dataset.
truncated_batch_size (int) – The maximum batch size.

Return type:

DpEvent

Returns:

A DpEvent object.