{"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/rezero-is-all-you-need-fast-convergence-at","title":"ReZero is All You Need: Fast Convergence at Large Depth","arxiv_id":"2003.04887","date":"2020-03-10","proceeding":null,"authors":["Thomas Bachlechner","Bodhisattwa Prasad Majumder","Huanru Henry Mao","Garrison W. Cottrell","Julian McAuley"],"abstract":"Deep networks often suffer from vanishing or exploding gradients due to inefficient signal propagation, leading to long training times or convergence difficulties. Various architecture designs, sophisticated residual-style networks, and initialization schemes have been shown to improve deep signal propagation. Recently, Pennington et al. used free probability theory to show that dynamical isometry plays an integral role in efficient deep learning. We show that the simplest architecture change of gating each residual connection using a single zero-initialized parameter satisfies initial dynamical isometry and outperforms more complex approaches. Although much simpler than its predecessors, this gate enables training thousands of fully connected layers with fast convergence and better test performance for ResNets trained on CIFAR-10. We apply this technique to language modeling and find that we can easily train 120-layer Transformers. When applied to 12 layer Transformers, it converges 56% faster on enwiki8.","url_abs":"https://arxiv.org/abs/2003.04887v2","url_pdf":"https://arxiv.org/pdf/2003.04887v2.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":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/majumderb/rezero","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/JunnYu/x-transformers-paddle","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/coolsunxu/rezero_tf2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/davidsvy/cosformer-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/fabio-deep/ReZero-ResNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/fengleifan/relinear","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/lucidrains/performer-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/lucidrains/reformer-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/lucidrains/routing-transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/sooheon/perceiver-jax","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":null},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/statsu1990/ReZero-Cifar100","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/tbachlechner/ReZero-examples","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"rezero-is-all-you-need-fast-convergence-at","repo_url":"https://github.com/EugenHotaj/pytorch-generative/blob/master/pytorch_generative/nn/utils.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"all","task_name":"All"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"rezero","method_name":"ReZero"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[{"slug":"rezero","name":"ReZero","full_name":"ReZero"}],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2003.04887","atlas_url":"https://app.syntology.ai/?focus=2003.04887","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2003.04887"}},"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/JunnYu/x-transformers-paddle","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sooheon/perceiver-jax","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lucidrains/routing-transformer","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lucidrains/performer-pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lucidrains/reformer-pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/davidsvy/cosformer-pytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/coolsunxu/rezero_tf2","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/statsu1990/ReZero-Cifar100","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/majumderb/rezero","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/fabio-deep/ReZero-ResNet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/fengleifan/relinear","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tbachlechner/ReZero-examples","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/EugenHotaj/pytorch-generative/blob/master/pytorch_generative/nn/utils.py","reach":null}],"summary":{"ran_draft_wrong":6},"by_repo_kind":{"listed":{"samples":6,"ran":6,"repositories":2}},"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":"09aa6bbcdc551903","entry":"PreActResNet18","repo":"statsu1990/ReZero-Cifar100","repo_kind":"listed","path":"src/model/rezero_preact_resnet.py","file_url":"https://github.com/statsu1990/ReZero-Cifar100/blob/HEAD/src/model/rezero_preact_resnet.py","link_basis":"first_harvest_node","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":"09aa6bbcdc551903"}},{"code_sha256_prefix":"228f5abb7a7bf1c7","entry":"PreActResNet34","repo":"statsu1990/ReZero-Cifar100","repo_kind":"listed","path":"src/model/rezero_preact_resnet.py","file_url":"https://github.com/statsu1990/ReZero-Cifar100/blob/HEAD/src/model/rezero_preact_resnet.py","link_basis":"first_harvest_node","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":"228f5abb7a7bf1c7"}},{"code_sha256_prefix":"347af4f109be8414","entry":"PreActResNet50","repo":"statsu1990/ReZero-Cifar100","repo_kind":"listed","path":"src/model/rezero_preact_resnet.py","file_url":"https://github.com/statsu1990/ReZero-Cifar100/blob/HEAD/src/model/rezero_preact_resnet.py","link_basis":"first_harvest_node","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":"347af4f109be8414"}},{"code_sha256_prefix":"ced375c8bbba0283","entry":"resnet20","repo":"fabio-deep/ReZero-ResNet","repo_kind":"listed","path":"src/resnet.py","file_url":"https://github.com/fabio-deep/ReZero-ResNet/blob/HEAD/src/resnet.py","link_basis":"first_harvest_node","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":"ced375c8bbba0283"}},{"code_sha256_prefix":"0062b25c7d99c79a","entry":"resnet32","repo":"fabio-deep/ReZero-ResNet","repo_kind":"listed","path":"src/resnet.py","file_url":"https://github.com/fabio-deep/ReZero-ResNet/blob/HEAD/src/resnet.py","link_basis":"first_harvest_node","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":"0062b25c7d99c79a"}},{"code_sha256_prefix":"7377861df3c47980","entry":"resnet44","repo":"fabio-deep/ReZero-ResNet","repo_kind":"listed","path":"src/resnet.py","file_url":"https://github.com/fabio-deep/ReZero-ResNet/blob/HEAD/src/resnet.py","link_basis":"first_harvest_node","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":"7377861df3c47980"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}