{"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/arxiv-2605-11838","title":"Gradient Clipping Beyond Vector Norms: A Spectral Approach for Matrix-Valued Parameters","arxiv_id":"2605.11838","date":"2026-05-12","proceeding":null,"authors":["Alexander Yukhimchuk","Mladen Kolar","Martin Takáč","Sayantan Choudhury"],"abstract":"Gradient clipping is a standard safeguard for training neural networks under noisy, heavy-tailed stochastic gradients; yet, most clipping rules treat all parameters as vectors and ignore the matrix structure of modern architectures. We show empirically that data outliers often amplify only a small number of leading singular values in layer-wise gradient matrices, while the rest of the spectrum remains largely unchanged. Motivated by this phenomenon, we propose spectral clipping, which stabilizes training by clamping singular values that exceed a threshold while preserving the singular directions. This framework generalizes classical gradient norm clipping and can be easily integrated into existing optimizers. We provide a convergence analysis for non-convex optimization with spectrally clipped SGD, yielding the optimal $\\mathcal{O}\\left(K^{\\frac{2 - 2α}{3α- 2}}\\right)$ rate for heavy-tailed noise. To minimize hyperparameter tuning, we introduce layer-wise adaptive thresholds based on moving averages or sliding-window quantiles of the top singular values. Finally, we develop efficient implementations that clip only the top $r$ singular values via randomized truncated SVD, avoiding full decompositions for large layers. We demonstrate competitive performance across synthetic heavy-tailed settings and neural network training tasks.","url_abs":"https://arxiv.org/abs/2605.11838","url_pdf":"https://arxiv.org/pdf/2605.11838","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2605.11838","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2605.11838"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/katcinskiy/spectral-clipping","reach":null}],"summary":{"ran_draft_wrong":1},"by_repo_kind":{"found_in_text":{"samples":1,"ran":1,"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":1,"samples":[{"code_sha256_prefix":"29f3e2314485d607","entry":"clip_singular_values_svd","repo":"katcinskiy/spectral-clipping","repo_kind":"found_in_text","path":"clipping.py","file_url":"https://github.com/katcinskiy/spectral-clipping/blob/HEAD/clipping.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":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"29f3e2314485d607"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}