{"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/git-re-basin-merging-models-modulo","title":"Git Re-Basin: Merging Models modulo Permutation Symmetries","arxiv_id":"2209.04836","date":"2022-09-11","proceeding":null,"authors":["Samuel K. Ainsworth","Jonathan Hayase","Siddhartha Srinivasa"],"abstract":"The success of deep learning is due in large part to our ability to solve certain massive non-convex optimization problems with relative ease. Though non-convex optimization is NP-hard, simple algorithms -- often variants of stochastic gradient descent -- exhibit surprising effectiveness in fitting large neural networks in practice. We argue that neural network loss landscapes often contain (nearly) a single basin after accounting for all possible permutation symmetries of hidden units a la Entezari et al. 2021. We introduce three algorithms to permute the units of one model to bring them into alignment with a reference model in order to merge the two models in weight space. This transformation produces a functionally equivalent set of weights that lie in an approximately convex basin near the reference model. Experimentally, we demonstrate the single basin phenomenon across a variety of model architectures and datasets, including the first (to our knowledge) demonstration of zero-barrier linear mode connectivity between independently trained ResNet models on CIFAR-10. Additionally, we identify intriguing phenomena relating model width and training time to mode connectivity. Finally, we discuss shortcomings of the linear mode connectivity hypothesis, including a counterexample to the single basin theory.","url_abs":"https://arxiv.org/abs/2209.04836v6","url_pdf":"https://arxiv.org/pdf/2209.04836v6.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":"git-re-basin-merging-models-modulo","repo_url":"https://github.com/samuela/git-re-basin","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"jax","reach":null},{"paper_slug":"git-re-basin-merging-models-modulo","repo_url":"https://github.com/gg-Dema/Git_merge","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"git-re-basin-merging-models-modulo","repo_url":"https://github.com/themrzmaster/git-re-basin-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"linear-mode-connectivity","task_name":"Linear Mode Connectivity"},{"task_slug":"re-basin","task_name":"Re-basin"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2209.04836","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2209.04836"}},"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/samuela/git-re-basin","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/themrzmaster/git-re-basin-pytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gg-Dema/Git_merge","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":5,"ran_honours":1},"by_repo_kind":{"official":{"samples":5,"ran":5,"repositories":1},"listed":{"samples":1,"ran":1,"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":"583f9780bdd00a45","entry":"conv3x3","repo":"themrzmaster/git-re-basin-pytorch","repo_kind":"listed","path":"models/resnet.py","file_url":"https://github.com/themrzmaster/git-re-basin-pytorch/blob/HEAD/models/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":"583f9780bdd00a45"}},{"code_sha256_prefix":"7a670f479fecf6dc","entry":"get_permuted_param","repo":"samuela/git-re-basin","repo_kind":"official","path":"src/weight_matching.py","file_url":"https://github.com/samuela/git-re-basin/blob/HEAD/src/weight_matching.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7a670f479fecf6dc"}},{"code_sha256_prefix":"1817b5650fd28afd","entry":"mlp_permutation_spec","repo":"samuela/git-re-basin","repo_kind":"official","path":"src/weight_matching.py","file_url":"https://github.com/samuela/git-re-basin/blob/HEAD/src/weight_matching.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1817b5650fd28afd"}},{"code_sha256_prefix":"602ebbf2f53305ff","entry":"permutation_spec_from_axes_to_perm","repo":"samuela/git-re-basin","repo_kind":"official","path":"src/weight_matching.py","file_url":"https://github.com/samuela/git-re-basin/blob/HEAD/src/weight_matching.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"602ebbf2f53305ff"}},{"code_sha256_prefix":"a6b09bc64f436279","entry":"plot_interp_acc","repo":"samuela/git-re-basin","repo_kind":"official","path":"src/cifar10_mlp_weight_matching.py","file_url":"https://github.com/samuela/git-re-basin/blob/HEAD/src/cifar10_mlp_weight_matching.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":"a6b09bc64f436279"}},{"code_sha256_prefix":"0494520c0147c1e8","entry":"plot_interp_loss","repo":"samuela/git-re-basin","repo_kind":"official","path":"src/cifar10_mlp_weight_matching.py","file_url":"https://github.com/samuela/git-re-basin/blob/HEAD/src/cifar10_mlp_weight_matching.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":"0494520c0147c1e8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}