{"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/exploiting-unlabeled-data-in-cnns-by-self","title":"Exploiting Unlabeled Data in CNNs by Self-supervised Learning to Rank","arxiv_id":"1902.06285","date":"2019-02-17","proceeding":null,"authors":["Xialei Liu","Joost Van de Weijer","Andrew D. Bagdanov"],"abstract":"For many applications the collection of labeled data is expensive laborious.\nExploitation of unlabeled data during training is thus a long pursued objective\nof machine learning. Self-supervised learning addresses this by positing an\nauxiliary task (different, but related to the supervised task) for which data\nis abundantly available. In this paper, we show how ranking can be used as a\nproxy task for some regression problems. As another contribution, we propose an\nefficient backpropagation technique for Siamese networks which prevents the\nredundant computation introduced by the multi-branch network architecture. We\napply our framework to two regression problems: Image Quality Assessment (IQA)\nand Crowd Counting. For both we show how to automatically generate ranked image\nsets from unlabeled data. Our results show that networks trained to regress to\nthe ground truth targets for labeled data and to simultaneously learn to rank\nunlabeled data obtain significantly better, state-of-the-art results for both\nIQA and crowd counting. In addition, we show that measuring network uncertainty\non the self-supervised proxy task is a good measure of informativeness of\nunlabeled data. This can be used to drive an algorithm for active learning and\nwe show that this reduces labeling effort by up to 50%.","url_abs":"http://arxiv.org/abs/1902.06285v1","url_pdf":"http://arxiv.org/pdf/1902.06285v1.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":"exploiting-unlabeled-data-in-cnns-by-self","repo_url":"https://github.com/LONG-9621/IQA_02","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"exploiting-unlabeled-data-in-cnns-by-self","repo_url":"https://github.com/xialeiliu/RankIQA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"active-learning","task_name":"Active Learning"},{"task_slug":"crowd-counting","task_name":"Crowd Counting"},{"task_slug":"image-quality-assessment","task_name":"Image Quality Assessment"},{"task_slug":"informativeness","task_name":"Informativeness"},{"task_slug":"learning-to-rank","task_name":"Learning-To-Rank"},{"task_slug":"self-supervised-learning","task_name":"Self-Supervised Learning"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1902.06285","atlas_url":"https://app.syntology.ai/?focus=1902.06285","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1902.06285"}},"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":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/LONG-9621/IQA_02","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/xialeiliu/RankIQA","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":2},"by_repo_kind":{"listed":{"samples":2,"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":"1cc5e3a6a2e74785","entry":"preprocess","repo":"xialeiliu/RankIQA","repo_kind":"listed","path":"src/data_layer/ft_layer_live.py","file_url":"https://github.com/xialeiliu/RankIQA/blob/HEAD/src/data_layer/ft_layer_live.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1cc5e3a6a2e74785"}},{"code_sha256_prefix":"4f33b8a7af1d4bfa","entry":"preprocess","repo":"xialeiliu/RankIQA","repo_kind":"listed","path":"src/data_layer/rank_layer_live.py","file_url":"https://github.com/xialeiliu/RankIQA/blob/HEAD/src/data_layer/rank_layer_live.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4f33b8a7af1d4bfa"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}