flip.nvflare.components.custom_percentile_privacy

Classes

PercentilePrivacy

Implementation of "largest percentile to share" privacy preserving policy.

Module Contents

class flip.nvflare.components.custom_percentile_privacy.PercentilePrivacy(percentile=10, gamma=0.01, data_kinds: List[str] | None = None, off: bool = False)[source]

Bases: nvflare.apis.dxo_filter.DXOFilter

Implementation of “largest percentile to share” privacy preserving policy.

Shokri and Shmatikov, Privacy-preserving deep learning, CCS ‘15

Parameters:

  • percentile (int, optional) – Only abs diff greater than this percentile is updated. Allowed range 0..100. Defaults to 10.

  • gamma (float, optional) – The upper limit to truncate abs values of weight diff. Defaults to 0.01.

  • 0. (Any weight diff with abs<gamma will become)

  • data_kinds – kinds of DXO to filter

  • off (bool, optional) – If True, the filter is turned off. Defaults to False.

percentile = 10[source]
gamma = 0.01[source]
off = False[source]
process_dxo(dxo: nvflare.apis.dxo.DXO, shareable: nvflare.apis.shareable.Shareable, fl_ctx: nvflare.apis.fl_context.FLContext) → None | nvflare.apis.dxo.DXO[source]

Compute the percentile on the abs delta_W.

Only share the params where absolute delta_W greater than the percentile value

Parameters:

  • dxo – information from client

  • shareable – that the dxo belongs to

  • fl_ctx – context provided by workflow

Returns: filtered dxo