{"url":"/method/adabound","slug":"adabound","name":"AdaBound","full_name":"AdaBound","full_name_withheld":false,"description_markdown":"**AdaBound** is a variant of the [Adam](https://paperswithcode.com/method/adabound) stochastic optimizer which is designed to be more robust to extreme learning rates. Dynamic bounds are employed on learning rates, where the lower and upper bound are initialized as zero and infinity respectively, and they both smoothly converge to a constant final step size. AdaBound can be regarded as an adaptive method at the beginning of training, and thereafter it gradually and smoothly transforms to [SGD](https://paperswithcode.com/method/sgd) (or with momentum) as the time step increases. \r\n\r\n$$ g\\_{t} = \\nabla{f}\\_{t}\\left(x\\_{t}\\right) $$\r\n\r\n$$ m\\_{t} = \\beta\\_{1t}m\\_{t-1} + \\left(1-\\beta\\_{1t}\\right)g\\_{t} $$\r\n\r\n$$ v\\_{t} = \\beta\\_{2}v\\_{t-1} + \\left(1-\\beta\\_{2}\\right)g\\_{t}^{2} \\text{ and } V\\_{t} = \\text{diag}\\left(v\\_{t}\\right) $$\r\n\r\n$$ \\hat{\\eta}\\_{t} = \\text{Clip}\\left(\\alpha/\\sqrt{V\\_{t}}, \\eta\\_{l}\\left(t\\right), \\eta\\_{u}\\left(t\\right)\\right) \\text{ and } \\eta\\_{t} = \\hat{\\eta}\\_{t}/\\sqrt{t} $$\r\n\r\n$$ x\\_{t+1} = \\Pi\\_{\\mathcal{F}, \\text{diag}\\left(\\eta\\_{t}^{-1}\\right)}\\left(x\\_{t} - \\eta\\_{t} \\odot m\\_{t} \\right) $$\r\n\r\nWhere $\\alpha$ is the initial step size, and $\\eta_{l}$ and $\\eta_{u}$ are the lower and upper bound functions respectively.","description_state":"present","introduced_year":null,"introduced_by":{"title":null,"paper":null,"first_author":null,"n_authors":0,"url_abs":null,"archive_paper_url":null},"source":{"url":"http://arxiv.org/abs/1902.09843v1","title":"Adaptive Gradient Methods with Dynamic Bound of Learning Rate","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/Luolc/AdaBound/blob/2e928c3007a2fc44af0e4c97e343e1fed6986e44/adabound/adabound.py#L6","code_snippet_url_on_a_code_host":true,"categories":[{"area":"General","area_id":"general","collection":"Stochastic Optimization","url":"/methods/category/stochastic-optimization","pwc_aliases":[]}],"n_papers_tagged":11,"archive_num_papers":null,"papers_newest_first":[{"paper":null,"title":"UAdam: Unified Adam-Type Algorithmic Framework for Non-Convex Stochastic Optimization","date":"2023-05-09","arxiv_id":"2305.05675","n_code_links":0,"syntology":null},{"paper":null,"title":"Unified Convergence Analysis for Adaptive Optimization with Moving Average Estimator","date":"2021-04-30","arxiv_id":"2104.14840","n_code_links":0,"syntology":null},{"paper":null,"title":"BERT-based Chinese Text Classification for Emergency Domain with a Novel Loss Function","date":"2021-04-09","arxiv_id":"2104.04197","n_code_links":0,"syntology":null},{"paper":null,"title":"Towards Better Generalization of Adaptive Gradient Methods","date":"2020-12-01","arxiv_id":null,"n_code_links":0,"syntology":null},{"paper":"/paper/fcm-rdpa-tsk-fuzzy-regression-model","title":"FCM-RDpA: TSK Fuzzy Regression Model Construction Using Fuzzy C-Means Clustering, Regularization, DropRule, and Powerball AdaBelief","date":"2020-11-30","arxiv_id":"2012.00060","n_code_links":2,"syntology":null},{"paper":"/paper/adaptive-learning-rates-with-maximum","title":"MaxVA: Fast Adaptation of Step Sizes by Maximizing Observed Variance of Gradients","date":"2020-06-21","arxiv_id":"2006.11918","n_code_links":1,"syntology":null},{"paper":null,"title":"MBGD-RDA Training and Rule Pruning for Concise TSK Fuzzy Regression Models","date":"2020-03-01","arxiv_id":"2003.00608","n_code_links":0,"syntology":null},{"paper":"/paper/an-adaptive-and-momental-bound-method-for","title":"An Adaptive and Momental Bound Method for Stochastic Learning","date":"2019-10-27","arxiv_id":"1910.12249","n_code_links":2,"syntology":{"ran":0,"of":2,"unverified":2,"pointer_only":0}},{"paper":"/paper/on-the-convergence-of-adabound-and-its","title":"On the Convergence of AdaBound and its Connection to SGD","date":"2019-08-13","arxiv_id":"1908.04457","n_code_links":2,"syntology":null},{"paper":"/paper/optimize-tsk-fuzzy-systems-for-big-data","title":"Optimize TSK Fuzzy Systems for Regression Problems: Mini-Batch Gradient Descent with Regularization, DropRule and AdaBound (MBGD-RDA)","date":"2019-03-26","arxiv_id":"1903.10951","n_code_links":1,"syntology":null},{"paper":"/paper/adaptive-gradient-methods-with-dynamic-bound","title":"Adaptive Gradient Methods with Dynamic Bound of Learning Rate","date":"2019-02-26","arxiv_id":"1902.09843","n_code_links":5,"syntology":null}],"papers_shown":11,"tasks":[{"task":"/task/stochastic-optimization","name":"Stochastic Optimization","papers":3},{"task":"/task/regression-1","name":"regression","papers":2},{"task":"/task/benchmarking","name":"Benchmarking","papers":1},{"task":"/task/bilevel-optimization","name":"Bilevel Optimization","papers":1},{"task":"/task/clustering","name":"Clustering","papers":1},{"task":"/task/classification","name":"General Classification","papers":1},{"task":"/task/image-classification","name":"Image Classification","papers":1},{"task":"/task/machine-translation","name":"Machine Translation","papers":1},{"task":"/task/management","name":"Management","papers":1},{"task":"/task/natural-language-understanding","name":"Natural Language Understanding","papers":1},{"task":"/task/text-categorization","name":"Text Categorization","papers":1},{"task":"/task/text-classification","name":"Text Classification","papers":1},{"task":"/task/translation","name":"Translation","papers":1},{"task":"/task/type","name":"Vocal Bursts Type Prediction","papers":1},{"task":"/task/text-classification-1","name":"text-classification","papers":1}],"tasks_shown":15,"n_tasks":15,"usage_by_year":[{"year":"2019","papers":4},{"year":"2020","papers":4},{"year":"2021","papers":2},{"year":"2023","papers":1}],"row_source":"embedded","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/adabound"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}