Continual Learning With Participation Privacy: An Auditable Buffering-Aggregation Recipe
This work provides a formal framework for participation privacy in continual learning, solving a known bottleneck for adaptive interaction in federated and streaming systems.
The paper addresses participation privacy in continual learning with single-edit neighboring user streams, where one insertion/deletion shifts all subsequent updates. It provides an auditable buffering-aggregation recipe that reduces single-edit streams to Hamming-style per-bin updates and proves trajectory-level (ε,δ)-DP with explicit privacy-latency trade-offs.
Modern federated and streaming learning systems often release intermediate models, so privacy must hold for the full trajectory under adaptive interaction. Motivated by participation privacy, we study single-edit neighboring user streams, where one insertion/deletion shifts all subsequent updates and defeats standard Hamming-neighbor continual-release analyses. We give an auditable modular recipe. A randomized buffering wrapper emits bins of size $[U,2U]$, reducing single-edit streams to a Hamming-style per-bin update stream with explicit backlog/delay guarantees, where $U$ is calibrated by the privacy parameters $(\varepsilon,δ)$. We then prove a certification theorem identifying when a non-adaptive Hamming-neighbor DP proof for a continual primitive lifts to adaptive inputs: the primitive must use fresh per-round randomness and have a stable one-round privacy profile under common adaptive context. Together, these ingredients yield trajectory-level $(\varepsilon,δ)$-DP for single-edit streams using standard primitives (e.g., tree prefix sums), with an explicit privacy--latency link via $U$.