{"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-2604-07328","title":"How to sketch a learning algorithm","arxiv_id":"2604.07328","date":"2026-04-08","proceeding":null,"authors":["Sam Gunn"],"abstract":"How does the choice of training data influence an AI model? This broad question is of central importance to interpretability, privacy, and basic science. At its technical core is the data deletion problem: after a reasonable amount of precomputation, quickly predict how the model would behave in a given situation if a given subset of training data had been excluded from the learning algorithm. We present a data deletion scheme capable of predicting model outputs with vanishing error $\\varepsilon$ and failure probability $δ$ in the deep learning setting. Our precomputation and prediction algorithms are only $\\tilde{O}(\\log(1/δ)/\\varepsilon^2)$ factors slower than regular training and inference, respectively. The storage requirements are those of $\\tilde{O}(\\log(1/δ)/\\varepsilon^2)$ models. Our proof is based on an assumption that we call stability. In contrast to the assumptions made by prior work, stability appears to be fully compatible with learning powerful AI models. In support of this, we show that stability is satisfied in a minimal set of experiments with microgpt. Our code is available at https://github.com/SamSpo1/microgpt-sketch. At a technical level, our work is based on a new method for locally sketching an arithmetic circuit by computing higher-order derivatives in random complex directions. Forward-mode automatic differentiation allows cheap computation of these derivatives.","url_abs":"https://arxiv.org/abs/2604.07328","url_pdf":"https://arxiv.org/pdf/2604.07328","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=2604.07328","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2604.07328"}},"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/SamSpo1/microgpt-sketch","reach":null}],"summary":{"unverified":7},"by_repo_kind":{"found_in_text":{"samples":7,"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":7,"samples":[{"code_sha256_prefix":"2aa12640eb87c06b","entry":"as_real","repo":"SamSpo1/microgpt-sketch","repo_kind":"found_in_text","path":"generate_fig1.py","file_url":"https://github.com/SamSpo1/microgpt-sketch/blob/HEAD/generate_fig1.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2aa12640eb87c06b"}},{"code_sha256_prefix":"3b3f08ca0e97a63e","entry":"as_real","repo":"SamSpo1/microgpt-sketch","repo_kind":"found_in_text","path":"results/fig1_data_to_tikz.py","file_url":"https://github.com/SamSpo1/microgpt-sketch/blob/HEAD/results/fig1_data_to_tikz.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3b3f08ca0e97a63e"}},{"code_sha256_prefix":"c5fd459bfa073471","entry":"coeff_name","repo":"SamSpo1/microgpt-sketch","repo_kind":"found_in_text","path":"results/fig2_data_to_tikz.py","file_url":"https://github.com/SamSpo1/microgpt-sketch/blob/HEAD/results/fig2_data_to_tikz.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c5fd459bfa073471"}},{"code_sha256_prefix":"beaaf9116a8a4a2f","entry":"color_style_for","repo":"SamSpo1/microgpt-sketch","repo_kind":"found_in_text","path":"results/fig3_data_to_tikz.py","file_url":"https://github.com/SamSpo1/microgpt-sketch/blob/HEAD/results/fig3_data_to_tikz.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"beaaf9116a8a4a2f"}},{"code_sha256_prefix":"ceafa205397edc0b","entry":"eval_truncated","repo":"SamSpo1/microgpt-sketch","repo_kind":"found_in_text","path":"results/fig2_data_to_tikz.py","file_url":"https://github.com/SamSpo1/microgpt-sketch/blob/HEAD/results/fig2_data_to_tikz.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ceafa205397edc0b"}},{"code_sha256_prefix":"c824d1e040c2f5d3","entry":"make_xticks","repo":"SamSpo1/microgpt-sketch","repo_kind":"found_in_text","path":"results/fig1_data_to_tikz.py","file_url":"https://github.com/SamSpo1/microgpt-sketch/blob/HEAD/results/fig1_data_to_tikz.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c824d1e040c2f5d3"}},{"code_sha256_prefix":"a4dda2021b193620","entry":"outward_int_bounds","repo":"SamSpo1/microgpt-sketch","repo_kind":"found_in_text","path":"results/fig1_data_to_tikz.py","file_url":"https://github.com/SamSpo1/microgpt-sketch/blob/HEAD/results/fig1_data_to_tikz.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a4dda2021b193620"}}]},"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}