{"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-2603-24695","title":"Amplified Patch-Level Differential Privacy for Free via Random Cropping","arxiv_id":"2603.24695","date":"2026-03-25","proceeding":null,"authors":["Kaan Durmaz","Jan Schuchardt","Sebastian Schmidt","Stephan Günnemann"],"abstract":"Random cropping is one of the most common data augmentation techniques in computer vision, yet the role of its inherent randomness in training differentially private machine learning models has thus far gone unexplored. We observe that when sensitive content in an image is spatially localized, such as a face or license plate, random cropping can probabilistically exclude that content from the model's input. This introduces a third source of stochasticity in differentially private training with stochastic gradient descent, in addition to gradient noise and minibatch sampling. This additional randomness amplifies differential privacy without requiring changes to model architecture or training procedure. We formalize this effect by introducing a patch-level neighboring relation for vision data and deriving tight privacy bounds for differentially private stochastic gradient descent (DP-SGD) when combined with random cropping. Our analysis quantifies the patch inclusion probability and shows how it composes with minibatch sampling to yield a lower effective sampling rate. Empirically, we validate that patch-level amplification improves the privacy-utility trade-off across multiple segmentation architectures and datasets. Our results demonstrate that aligning privacy accounting with domain structure and additional existing sources of randomness can yield stronger guarantees at no additional cost.","url_abs":"https://arxiv.org/abs/2603.24695","url_pdf":"https://arxiv.org/pdf/2603.24695","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":{"syntology_url":"https://syntology.ai/paper/2603.24695","atlas_url":"https://app.syntology.ai/?focus=2603.24695","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2603.24695"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/TUM-DAML/patch_level_dp","reach":null}],"summary":{"ran":4,"unverified":1},"by_repo_kind":{"found_in_text":{"samples":5,"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":5,"samples":[{"code_sha256_prefix":"462e49a02ad3f79d","entry":"dtype_safe","repo":"TUM-DAML/patch_level_dp","repo_kind":"found_in_text","path":"src/patch_level_dp/privacy/dataloader.py","file_url":"https://github.com/TUM-DAML/patch_level_dp/blob/HEAD/src/patch_level_dp/privacy/dataloader.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"462e49a02ad3f79d"}},{"code_sha256_prefix":"8d5d894aeffaa8e4","entry":"generate_circle_pixels","repo":"TUM-DAML/patch_level_dp","repo_kind":"found_in_text","path":"src/patch_level_dp/privacy/calculations.py","file_url":"https://github.com/TUM-DAML/patch_level_dp/blob/HEAD/src/patch_level_dp/privacy/calculations.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8d5d894aeffaa8e4"}},{"code_sha256_prefix":"4d31b5b2ddd4f7e9","entry":"generate_rectangle_pixels","repo":"TUM-DAML/patch_level_dp","repo_kind":"found_in_text","path":"src/patch_level_dp/privacy/calculations.py","file_url":"https://github.com/TUM-DAML/patch_level_dp/blob/HEAD/src/patch_level_dp/privacy/calculations.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4d31b5b2ddd4f7e9"}},{"code_sha256_prefix":"a244707168006dd0","entry":"shape_safe","repo":"TUM-DAML/patch_level_dp","repo_kind":"found_in_text","path":"src/patch_level_dp/privacy/dataloader.py","file_url":"https://github.com/TUM-DAML/patch_level_dp/blob/HEAD/src/patch_level_dp/privacy/dataloader.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a244707168006dd0"}},{"code_sha256_prefix":"c2f5f5309d0f6648","entry":"collate","repo":"TUM-DAML/patch_level_dp","repo_kind":"found_in_text","path":"src/patch_level_dp/privacy/dataloader.py","file_url":"https://github.com/TUM-DAML/patch_level_dp/blob/HEAD/src/patch_level_dp/privacy/dataloader.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":"c2f5f5309d0f6648"}}]},"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}