paper-with-me

홈 › Papers

DP-AdamBC: Your DP-Adam Is Actually DP-SGD (Unless You Apply Bias Correction)

2023-12-21 · Qiaoyue Tang, Frederick Shpilevskiy, Mathias Lécuyer

The Adam optimizer is a popular choice in contemporary deep learning, due to its strong empirical performance. However we observe that in privacy sensitive scenarios, the traditional use of Differential Privacy (DP) with the Adam optimizer leads to sub-optimal performance on several tasks. We find that this performance degradation is due to a DP bias in Adam's second moment estimator, introduced by the addition of independent noise in the gradient computation to enforce DP guarantees. This DP bias leads to a different scaling for low variance parameter updates, that is inconsistent with the behavior of non-private Adam. We propose DP-AdamBC, an optimization algorithm which removes the bias in the second moment estimation and retrieves the expected behaviour of Adam. Empirically, DP-AdamBC significantly improves the optimization performance of DP-Adam by up to 3.5% in final accuracy in image, text, and graph node classification tasks.

📄 PDF Abstract BibTeX arXiv:2312.14334

Code (1)

ubc-systopia/DP-AdamBC 공식 구현 jax

Tasks

Node Classification

Methods 이 논문이 사용한 방법론

Adam 설명 없음

Similar Papers 제목 키워드 기반

DP-AdamW: Investigating Decoupled Weight Decay and Bias Correction in Private Deep Learning

2025-11-11 · Jay Chooi, Kevin Cong, Russell Li, Lillian Sun arxiv

As deep learning methods increasingly utilize sensitive data on a widespread scale, differential privacy (DP) offers formal guarantees to protect against information leakage during model training. A significant challenge…

Image ClassificationNode ClassificationText Classification

Is your batch size the problem? Revisiting the Adam-SGD gap in language modeling

2025-06-14 · Teodora Srećković, Jonas Geiping, Antonio Orvieto

Adam is known to perform significantly better than Stochastic Gradient Descent (SGD) in language models, a phenomenon for which a number of explanations have been proposed. In this work, we revisit this "optimizer gap" t…

Language ModelingLanguage Modelling

On the Performance of Differentially Private Optimization with Heavy-Tail Class Imbalance

2025-07-14 · Qiaoyue Tang, Alain Zhiyanov, Mathias Lécuyer arxiv

In this work, we analyze the optimization behaviour of common private learning optimization algorithms under heavy-tail class imbalanced distribution. We show that, in a stylized model, optimizing with Gradient Descent w…

ADOPT: Modified Adam Can Converge with Any $β_2$ with the Optimal Rate

2024-11-05 · Shohei Taniguchi, Keno Harada, Gouki Minegishi, Yuta Oshima 외

Adam is one of the most popular optimization algorithms in deep learning. However, it is known that Adam does not converge in theory unless choosing a hyperparameter, i.e., $\beta_2$, in a problem-dependent manner. There…

Deep Reinforcement Learningimage-classificationImage Classification

Does Adam Converge and When?

2022-01-17 · ICLR Track Blog 2022 5 · Anonymous

In this blog post, we revisit the (non-)convergence behavior of Adam. Especially, we briefly review the non-convergence results by Reddi et al'19 and the convergence results by Shi et al.'20. Their results take important…