Differentially Private Adaptive Optimization with Delayed PreconditionersDownload PDF

Published: 01 Feb 2023, Last Modified: 12 Mar 2024ICLR 2023 posterReaders: Everyone
Keywords: adaptive optimization, differential privacy
Abstract: Privacy costs may negate the benefits of using adaptive optimizers in differentially private model training. Prior works typically address this issue by using auxiliary information (e.g., public data) to boost the effectiveness of adaptive optimization. In this work, we explore techniques to estimate and efficiently adapt to gradient geometry in private adaptive optimization without auxiliary data. Motivated by the observation that adaptive methods can tolerate stale preconditioners, we propose differentially private adaptive training with delayed preconditioners (DP^2), a simple method that constructs delayed but less noisy preconditioners to better realize the benefits of adaptivity. Theoretically, we provide convergence guarantees for our method for both convex and non-convex problems, and analyze trade-offs between delay and privacy noise reduction. Empirically, we explore DP^2 across several real-world datasets, demonstrating that it can improve convergence speed by as much as 4× relative to non-adaptive baselines and match the performance of state-of-the-art optimization methods that require auxiliary data.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
Submission Guidelines: Yes
Please Choose The Closest Area That Your Submission Falls Into: Optimization (eg, convex and non-convex optimization)
TL;DR: We propose a private adaptive optimization framework that constructs delayed but less noisy preconditioners, yielding improved privacy/utility trade-offs without the need to access auxiliary data.
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/arxiv:2212.00309/code)
15 Replies

Loading