paper-with-me

홈 › Papers

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 steps forward to understand Adam better. However, the convergence analysis by Shi et al.'20 requires $\beta_1$ to be either 0 or small enough ($\beta_1$ is the momentum hyperparameter in Adam). Is this a reasonable requirement? If not, how large is the gap between theory and practice? In this blog, we will discuss these questions from multiple different perspectives. We will show that the gap is actually non-negligible, and the discussion on the convergence of Adam is far from being concluded.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Adam 설명 없음

Similar Papers 제목 키워드 기반

Provable Adaptivity of Adam under Non-uniform Smoothness

2022-08-21 · Bohan Wang, Yushun Zhang, Huishuai Zhang, Qi Meng 외

Adam is widely adopted in practical applications due to its fast convergence. However, its theoretical analysis is still far from satisfactory. Existing convergence analyses for Adam rely on the bounded smoothness assump…

Attribute

Adam Converges Without Any Modification On Update Rules

2026-03-02 · Yushun Zhang, Bingran Li, Congliang Chen, Zhi-Quan Luo 외 arxiv

Adam is the default algorithm for training neural networks, including large language models (LLMs). However, \citet{reddi2019convergence} provided an example that Adam diverges, raising concerns for its deployment in AI …

A Simple Convergence Proof of Adam and Adagrad

2020-03-05 · Alexandre Défossez, Léon Bottou, Francis Bach, Nicolas Usunier

We provide a simple proof of convergence covering both the Adam and Adagrad adaptive optimization algorithms when applied to smooth (possibly non-convex) objective functions with bounded gradients. We show that in expect…

AdamL: A fast adaptive gradient method incorporating loss function

2023-12-23 · Lu Xia, Stefano Massei

Adaptive first-order optimizers are fundamental tools in deep learning, although they may suffer from poor generalization due to the nonuniform gradient scaling. In this work, we propose AdamL, a novel variant of the Ada…

Adam Can Converge Without Any Modification On Update Rules

2022-08-20 · Yushun Zhang, Congliang Chen, Naichen Shi, Ruoyu Sun 외

Ever since Reddi et al. 2018 pointed out the divergence issue of Adam, many new variants have been designed to obtain convergence. However, vanilla Adam remains exceptionally popular and it works well in practice. Why is…