{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/method/graddrop","slug":"graddrop","name":"GradDrop","full_name":"Gradient Sign Dropout","full_name_withheld":false,"description_markdown":"**GradDrop**, or **Gradient Sign Dropout**, is a probabilistic masking procedure which samples gradients at an activation layer based on their level of consistency. It is applied as a layer in any standard network forward pass, usually on the final layer before the prediction head to save on compute overhead and maximize benefits during backpropagation. Below, we develop the GradDrop formalism. Throughout, o denotes elementwise multiplication after any necessary tiling operations (if any) are completed.\r\nTo implement GradDrop, we first define the Gradient Positive Sign Purity, $\\mathcal{P}$, as\r\n\r\n$$\r\n\\mathcal{P}=\\frac{1}{2}\\left(1+\\frac{\\sum\\_{i} \\nabla L_\\{i}}{\\sum\\_{i}\\left|\\nabla L\\_{i}\\right|}\\right)\r\n$$\r\n\r\n$\\mathcal{P}$ is bounded by $[0,1] .$ For multiple gradient values $\\nabla\\_{a} L\\_{i}$ at some scalar $a$, we see that $\\mathcal{P}=0$ if $\\nabla_{a} L\\_{i}<0 $ $\\forall i$, while $\\mathcal{P}=1$ if $\\nabla\\_{a} L\\_{i}>0$ $\\forall i $. Thus, $\\mathcal{P}$ is a measure of how many positive gradients are present at any given value. We then form a mask for each gradient $\\mathcal{M}\\_{i}$ as follows:\r\n\r\n$$\r\n\\mathcal{M}\\_{i}=\\mathcal{I}[f(\\mathcal{P})>U] \\circ \\mathcal{I}\\left[\\nabla L\\_{i}>0\\right]+\\mathcal{I}[f(\\mathcal{P})<U] \\circ \\mathcal{I}\\left[\\nabla L\\_{i}<0\\right]\r\n$$\r\n\r\nfor $\\mathcal{I}$ the standard indicator function and $f$ some monotonically increasing function (often just the identity) that maps $[0,1] \\mapsto[0,1]$ and is odd around $(0.5,0.5)$. $U$ is a tensor composed of i.i.d $U(0,1)$ random variables. The $\\mathcal{M}\\_{i}$ is then used to produce a final gradient $\\sum \\mathcal{M}\\_{i} \\nabla L\\_{i}$","description_state":"present","introduced_year":null,"introduced_by":{"title":"Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout","paper":"/paper/just-pick-a-sign-optimizing-deep-multitask","first_author":"Zhao Chen","n_authors":7,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/just-pick-a-sign-optimizing-deep-multitask"},"source":{"url":"https://arxiv.org/abs/2010.06808v1","title":"Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"General","area_id":"general","collection":"Regularization","url":"/methods/category/regularization","pwc_aliases":[]}],"n_papers_tagged":3,"archive_num_papers":3,"papers_newest_first":[{"paper":null,"slug":"gradient-sparsification-for-masked-fine","title":"Gradient Sparsification For Masked Fine-Tuning of Transformers","date":"2023-07-19","arxiv_id":"2307.10098","n_code_links":0,"syntology":null},{"paper":null,"slug":"gradient-sparsification-for-emph-masked-fine","title":"Gradient Sparsification For \\emph{Masked Fine-Tuning} of Transformers","date":"2021-11-16","arxiv_id":null,"n_code_links":0,"syntology":null},{"paper":"/paper/just-pick-a-sign-optimizing-deep-multitask","slug":"just-pick-a-sign-optimizing-deep-multitask","title":"Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout","date":"2020-10-14","arxiv_id":"2010.06808","n_code_links":2,"syntology":{"ran":3,"of":6,"n_ran_checked":3,"n_instrument":0,"unverified":3,"pointer_only":6,"phrase":"3 ran (of which 1 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 1 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified","official":{"repos":["tensorflow/lingvo"],"state":"community repositories only","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["listed"]}}}],"papers_shown":3,"papers_order":"date (newest first), then slug","paper_counts":{"archive_papers_tagged":3,"with_a_code_link":1,"where_syntology_ran_a_sample":1,"not_listed_spam_title":0,"listed":3,"listed_where_code_ran":1,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":1,"every_run_a_failure_of_syntologys_instrument":0,"listed_with_a_run_with_no_instrument_failure":1,"listed_every_run_a_failure_of_syntologys_instrument":0,"filter":{"states":["a run with no instrument failure","any run, instrument failures included"],"default":"a run with no instrument failure","note":"on the 'only where code ran' pages the default hides, in the browser, the rows where every run was a failure of Syntology's instrument; the second state shows them again. Rows are hidden, never re-ordered; these twins list every row"}},"definition":"distinct papers the archive tags; 'where Syntology ran a sample' counts papers with at least one harvested sample that ran, which is not a correctness claim"},"papers_where_code_ran":1,"papers_where_code_ran_first_page":"/method/graddrop/papers/ran/1","papers_where_code_ran_caption":"We ran code from the paper's repository; we did not isolate this method inside it.","pages":[{"order":"ran","page":1,"of":1,"url":"/method/graddrop/papers/ran/1","json":"/method/graddrop/papers/ran/1.json","rows":1}],"tasks":[{"task":"/task/transfer-learning","name":"Transfer Learning","papers":3},{"task":"/task/xlm-r","name":"XLM-R","papers":1}],"tasks_shown":2,"n_tasks":2,"usage_by_year":[{"year":"2020","papers":1},{"year":"2021","papers":1},{"year":"2023","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/graddrop"},"syntology_read_at":"2026-10-05T00:36:11+00:00","syntology":{"source":"Syntology's have check for this name (the free coverage lookup), read from Syntology's graph","have_url":"https://syntology.ai/api/graph/have?x=GradDrop","read_at":"2026-10-05T00:57:11+00:00","query":"GradDrop","kind":"method","have":"unknown_name","tier":null,"verified_implementation":null,"level":null,"level_meaning":"the served verification level, per answer; never summed across methods","resolved_to":null,"resolution":null,"resolved_name_differs":false,"absence_asserted":false,"absence_unchecked":null,"no_code_is_about_this_method":false,"subject":null,"archive_paper_arxiv_id":"2010.06808","repositories":null,"retrieve_with":null,"name_competition":{"verdict":"none","may_be_wrong_sense":false,"bearers_found":0,"served":null,"leading_other_bearer":null,"n_other_bearers_listed":0,"note":"Nothing in this corpus is named 'GradDrop' -- no method node, no paper announcing it in its title, no expanded form declaring it as an acronym."},"name_competition_text":null,"near_matches":[],"code_nearby":[],"code_nearby_caption":null,"origin_paper":null,"line":"name not resolved (unknown_name). The lookup matched no method, paper or venue by this name; that asserts nothing about whether code exists.","level_legend":null},"syntology_have_state":"loaded","record_sha256":"d3b29190bb0fe9b01d37018f1ac2cbd037f9b9432a844a091b9ace016d5466ed","record_changed_at":"2026-10-05","record_changed_at_basis":"changed"}