AI Paper Atlas

Research map / Optimization / Optimizers

Towards Practical Adam: Non-Convexity, Convergence Theory, and Mini-Batch Acceleration

Congliang Chen, Li Shen, Fangyu Zou, Wei Liu

JMLR 2022 · Optimization › Optimizers · approach: Adam convergence and adaptive learning rates · 19 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase exponent r of theta_t (r=0 to 1) in Generic Adam (ablation)Generic Adam · 1-D counterexampleconvergenceImproved
use theta_t increasing to 1 (r>0) instead of constant theta (r=0, original Adam) (ablation)Generic Adam · LeNet on MNIST and ResNet-18 on CIFAR-100training loss and test accuracyImproved
increase batch size for mini-batch SGD and SGD-momentum (ablation)SGD · Quadratic minimizationoptimal learning rate: 0.005 -> 0.04Mixed
increase batch size for mini-batch Adam (ablation)Adam · Quadratic minimizationoptimal learning rate: between 0.02 to 0.04No clear effect
increase batch size (32, 64, 128) for mini-batch Adam (ablation)Adam · MNIST and CIFAR-100training lossImproved
increase batch size (32, 64, 128) for mini-batch Adam (ablation)Adam · MNIST and CIFAR-100test accuracy and test lossNo clear effect
use adaptive methods (Adam, AMSGrad) instead of SGD-MTransformer XL · WikiText-103training lossImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy