{"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/proxsarah-an-efficient-algorithmic-framework","title":"ProxSARAH: An Efficient Algorithmic Framework for Stochastic Composite Nonconvex Optimization","arxiv_id":"1902.05679","date":"2019-02-15","proceeding":null,"authors":["Nhan H. Pham","Lam M. Nguyen","Dzung T. Phan","Quoc Tran-Dinh"],"abstract":"We propose a new stochastic first-order algorithmic framework to solve\nstochastic composite nonconvex optimization problems that covers both\nfinite-sum and expectation settings. Our algorithms rely on the SARAH estimator\nintroduced in (Nguyen et al, 2017) and consist of two steps: a proximal\ngradient and an averaging step making them different from existing nonconvex\nproximal-type algorithms. The algorithms only require an average smoothness\nassumption of the nonconvex objective term and additional bounded variance\nassumption if applied to expectation problems. They work with both constant and\nadaptive step-sizes, while allowing single sample and mini-batches. In all\nthese cases, we prove that our algorithms can achieve the best-known complexity\nbounds. One key step of our methods is new constant and adaptive step-sizes\nthat help to achieve desired complexity bounds while improving practical\nperformance. Our constant step-size is much larger than existing methods\nincluding proximal SVRG schemes in the single sample case. We also specify the\nalgorithm to the non-composite case that covers existing state-of-the-arts in\nterms of complexity bounds. Our update also allows one to trade-off between\nstep-sizes and mini-batch sizes to improve performance. We test the proposed\nalgorithms on two composite nonconvex problems and neural networks using\nseveral well-known datasets.","url_abs":"http://arxiv.org/abs/1902.05679v2","url_pdf":"http://arxiv.org/pdf/1902.05679v2.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":"proxsarah-an-efficient-algorithmic-framework","repo_url":"https://github.com/unc-optimization/StochasticProximalMethods","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1902.05679","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1902.05679"}},"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/unc-optimization/StochasticProximalMethods","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"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":"851bd502f2d7c6e7","entry":"prox_sarah","repo":"unc-optimization/StochasticProximalMethods","repo_kind":"official","path":"python_src/method_ProxSARAH.py","file_url":"https://github.com/unc-optimization/StochasticProximalMethods/blob/HEAD/python_src/method_ProxSARAH.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"851bd502f2d7c6e7"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}