{"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/dada-differentiable-automatic-data","title":"DADA: Differentiable Automatic Data Augmentation","arxiv_id":"2003.03780","date":"2020-03-08","proceeding":"ECCV 2020 8","authors":["Yonggang Li","Guosheng Hu","Yongtao Wang","Timothy Hospedales","Neil M. Robertson","Yongxin Yang"],"abstract":"Data augmentation (DA) techniques aim to increase data variability, and thus train deep networks with better generalisation. The pioneering AutoAugment automated the search for optimal DA policies with reinforcement learning. However, AutoAugment is extremely computationally expensive, limiting its wide applicability. Followup works such as Population Based Augmentation (PBA) and Fast AutoAugment improved efficiency, but their optimization speed remains a bottleneck. In this paper, we propose Differentiable Automatic Data Augmentation (DADA) which dramatically reduces the cost. DADA relaxes the discrete DA policy selection to a differentiable optimization problem via Gumbel-Softmax. In addition, we introduce an unbiased gradient estimator, RELAX, leading to an efficient and effective one-pass optimization strategy to learn an efficient and accurate DA policy. We conduct extensive experiments on CIFAR-10, CIFAR-100, SVHN, and ImageNet datasets. Furthermore, we demonstrate the value of Auto DA in pre-training for downstream detection problems. Results show our DADA is at least one order of magnitude faster than the state-of-the-art while achieving very comparable accuracy. The code is available at https://github.com/VDIGPKU/DADA.","url_abs":"https://arxiv.org/abs/2003.03780v3","url_pdf":"https://arxiv.org/pdf/2003.03780v3.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":"dada-differentiable-automatic-data","repo_url":"https://github.com/VDIGPKU/DADA","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[{"method_slug":"autoaugment","method_name":"AutoAugment"},{"method_slug":"fast-autoaugment","method_name":"Fast AutoAugment"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"population-based-augmentation","method_name":"Population Based Augmentation"},{"method_slug":"population-based-training","method_name":"Population Based Training"},{"method_slug":"speed","method_name":"SPEED"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/data-augmentation-on-imagenet","task":"Data Augmentation","dataset":"ImageNet","model":"ResNet-50 (DADA)","rank_in_archive_order":15,"of":17,"metrics":{"Accuracy (%)":"77.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2003.03780","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2003.03780"}},"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/VDIGPKU/DADA","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":3,"ran":2,"unverified":4},"by_repo_kind":{"official":{"samples":9,"ran":5,"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":"fac5364e2f53c6db","entry":"conv3x3","repo":"VDIGPKU/DADA","repo_kind":"official","path":"search_gumbel/networks/resnet.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/search_gumbel/networks/resnet.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fac5364e2f53c6db"}},{"code_sha256_prefix":"00e569acd6b45ef0","entry":"conv3x3","repo":"VDIGPKU/DADA","repo_kind":"official","path":"search_gumbel/networks/wideresnet.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/search_gumbel/networks/wideresnet.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"00e569acd6b45ef0"}},{"code_sha256_prefix":"03a0be4fb7381bd2","entry":"conv3x3","repo":"VDIGPKU/DADA","repo_kind":"official","path":"search_gumbel/networks/pyramidnet.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/search_gumbel/networks/pyramidnet.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"03a0be4fb7381bd2"}},{"code_sha256_prefix":"40d3c39a403ae1f7","entry":"float_parameter","repo":"VDIGPKU/DADA","repo_kind":"official","path":"fast-autoaugment/archive.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/fast-autoaugment/archive.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"40d3c39a403ae1f7"}},{"code_sha256_prefix":"7900a3a6deef908b","entry":"int_parameter","repo":"VDIGPKU/DADA","repo_kind":"official","path":"fast-autoaugment/archive.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/fast-autoaugment/archive.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7900a3a6deef908b"}},{"code_sha256_prefix":"ba7d4c39d8ffaaa2","entry":"num_class","repo":"VDIGPKU/DADA","repo_kind":"official","path":"search_gumbel/dataset.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/search_gumbel/dataset.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":"ba7d4c39d8ffaaa2"}},{"code_sha256_prefix":"10d196b211843339","entry":"parse_devkit","repo":"VDIGPKU/DADA","repo_kind":"official","path":"search_gumbel/imagenet.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/search_gumbel/imagenet.py","link_basis":"harvester_set","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":"10d196b211843339"}},{"code_sha256_prefix":"97bfaa2cce837d67","entry":"parse_meta","repo":"VDIGPKU/DADA","repo_kind":"official","path":"search_gumbel/imagenet.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/search_gumbel/imagenet.py","link_basis":"harvester_set","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":"97bfaa2cce837d67"}},{"code_sha256_prefix":"e7825a1705689714","entry":"parse_val_groundtruth","repo":"VDIGPKU/DADA","repo_kind":"official","path":"search_gumbel/imagenet.py","file_url":"https://github.com/VDIGPKU/DADA/blob/HEAD/search_gumbel/imagenet.py","link_basis":"harvester_set","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":"e7825a1705689714"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}