{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/slowing-down-the-weight-norm-increase-in","title":"AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights","arxiv_id":"2006.08217","date":"2020-06-15","proceeding":"ICLR 2021 1","authors":["Byeongho Heo","Sanghyuk Chun","Seong Joon Oh","Dongyoon Han","Sangdoo Yun","Gyuwan Kim","Youngjung Uh","Jung-Woo Ha"],"abstract":"Normalization techniques are a boon for modern deep learning. They let weights converge more quickly with often better generalization performances. It has been argued that the normalization-induced scale invariance among the weights provides an advantageous ground for gradient descent (GD) optimizers: the effective step sizes are automatically reduced over time, stabilizing the overall training procedure. It is often overlooked, however, that the additional introduction of momentum in GD optimizers results in a far more rapid reduction in effective step sizes for scale-invariant weights, a phenomenon that has not yet been studied and may have caused unwanted side effects in the current practice. This is a crucial issue because arguably the vast majority of modern deep neural networks consist of (1) momentum-based GD (e.g. SGD or Adam) and (2) scale-invariant parameters. In this paper, we verify that the widely-adopted combination of the two ingredients lead to the premature decay of effective step sizes and sub-optimal model performances. We propose a simple and effective remedy, SGDP and AdamP: get rid of the radial component, or the norm-increasing direction, at each optimizer step. Because of the scale invariance, this modification only alters the effective step sizes without changing the effective update directions, thus enjoying the original convergence properties of GD optimizers. Given the ubiquity of momentum GD and scale invariance in machine learning, we have evaluated our methods against the baselines on 13 benchmarks. They range from vision tasks like classification (e.g. ImageNet), retrieval (e.g. CUB and SOP), and detection (e.g. COCO) to language modelling (e.g. WikiText) and audio classification (e.g. DCASE) tasks. We verify that our solution brings about uniform gains in those benchmarks. Source code is available at https://github.com/clovaai/AdamP.","url_abs":"https://arxiv.org/abs/2006.08217v3","url_pdf":"https://arxiv.org/pdf/2006.08217v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"slowing-down-the-weight-norm-increase-in","repo_url":"https://github.com/clovaai/AdamP","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"slowing-down-the-weight-norm-increase-in","repo_url":"https://github.com/rwightman/pytorch-image-models","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"slowing-down-the-weight-norm-increase-in","repo_url":"https://github.com/namakemono/kaggle-alaska2-image-steganalysis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"paper_slug":"slowing-down-the-weight-norm-increase-in","repo_url":"https://github.com/taki0112/AdamP-Tensorflow","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"audio-classification","task_name":"Audio Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"sgd","method_name":"SGD"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2006.08217","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2006.08217"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/rwightman/pytorch-image-models","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/clovaai/adamp","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/taki0112/AdamP-Tensorflow","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/namakemono/kaggle-alaska2-image-steganalysis","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/clovaai/AdamP","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":1,"ran_draft_wrong":2,"ran_honours":1},"by_repo_kind":{"named_in_paper":{"samples":4,"ran":4,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"7dee237780592a3a","entry":"AdamP","repo":"rwightman/pytorch-image-models","repo_kind":"named_in_paper","path":"timm/optim/adamp.py","file_url":"https://github.com/rwightman/pytorch-image-models/blob/HEAD/timm/optim/adamp.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7dee237780592a3a"}},{"code_sha256_prefix":"34d91c0e73aa4bbc","entry":"_channel_view","repo":"rwightman/pytorch-image-models","repo_kind":"named_in_paper","path":"timm/optim/adamp.py","file_url":"https://github.com/rwightman/pytorch-image-models/blob/HEAD/timm/optim/adamp.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"34d91c0e73aa4bbc"}},{"code_sha256_prefix":"18af39bdf46679d2","entry":"_layer_view","repo":"rwightman/pytorch-image-models","repo_kind":"named_in_paper","path":"timm/optim/adamp.py","file_url":"https://github.com/rwightman/pytorch-image-models/blob/HEAD/timm/optim/adamp.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"18af39bdf46679d2"}},{"code_sha256_prefix":"de5acc29c73a2126","entry":"projection","repo":"rwightman/pytorch-image-models","repo_kind":"named_in_paper","path":"timm/optim/adamp.py","file_url":"https://github.com/rwightman/pytorch-image-models/blob/HEAD/timm/optim/adamp.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"de5acc29c73a2126"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}