{"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/deep-learning-without-shortcuts-shaping-the-1","title":"Deep Learning without Shortcuts: Shaping the Kernel with Tailored Rectifiers","arxiv_id":"2203.08120","date":"2022-03-15","proceeding":"ICLR 2022 4","authors":["Guodong Zhang","Aleksandar Botev","James Martens"],"abstract":"Training very deep neural networks is still an extremely challenging task. The common solution is to use shortcut connections and normalization layers, which are both crucial ingredients in the popular ResNet architecture. However, there is strong evidence to suggest that ResNets behave more like ensembles of shallower networks than truly deep ones. Recently, it was shown that deep vanilla networks (i.e. networks without normalization layers or shortcut connections) can be trained as fast as ResNets by applying certain transformations to their activation functions. However, this method (called Deep Kernel Shaping) isn't fully compatible with ReLUs, and produces networks that overfit significantly more than ResNets on ImageNet. In this work, we rectify this situation by developing a new type of transformation that is fully compatible with a variant of ReLUs -- Leaky ReLUs. We show in experiments that our method, which introduces negligible extra computational cost, achieves validation accuracies with deep vanilla networks that are competitive with ResNets (of the same width/depth), and significantly higher than those obtained with the Edge of Chaos (EOC) method. And unlike with EOC, the validation accuracies we obtain do not get worse with depth.","url_abs":"https://arxiv.org/abs/2203.08120v1","url_pdf":"https://arxiv.org/pdf/2203.08120v1.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":"deep-learning-without-shortcuts-shaping-the-1","repo_url":"https://github.com/deepmind/dks","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"jax","reach":null}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"}],"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=2203.08120","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2203.08120"}},"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":"deterministic:regex_extraction","url":"https://github.com/deepmind/dks","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/deepmind/kfac_jax","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":4,"ran_honours":7,"ran_violates":1,"ran_draft_wrong":2,"unverified":10},"by_repo_kind":{"official":{"samples":18,"ran":11,"repositories":1},"found_in_text":{"samples":6,"ran":3,"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":"b803f9baa24bfd4f","entry":"_bentid","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b803f9baa24bfd4f"}},{"code_sha256_prefix":"4a9bd4aba951f2ee","entry":"_calc_c_curv","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4a9bd4aba951f2ee"}},{"code_sha256_prefix":"af6685abaca2e383","entry":"_calc_c_slope","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"af6685abaca2e383"}},{"code_sha256_prefix":"66703fda3e8cfbe3","entry":"_elu","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"66703fda3e8cfbe3"}},{"code_sha256_prefix":"b4a306c2c7f07155","entry":"_estimate_2d_gaussian_mean","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.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":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b4a306c2c7f07155"}},{"code_sha256_prefix":"c8c394e2c6a1a2ce","entry":"_gelu","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c8c394e2c6a1a2ce"}},{"code_sha256_prefix":"45c504b4c4b5cb39","entry":"_gelu_exact","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"45c504b4c4b5cb39"}},{"code_sha256_prefix":"fecba24c85a8383c","entry":"_get_roots_cache_file","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.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":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"fecba24c85a8383c"}},{"code_sha256_prefix":"8e18417d18243196","entry":"_softplus","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8e18417d18243196"}},{"code_sha256_prefix":"2c59e2c9e3970f5a","entry":"_swish","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2c59e2c9e3970f5a"}},{"code_sha256_prefix":"5afc010dc0e842f8","entry":"_transform_activation","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"5afc010dc0e842f8"}},{"code_sha256_prefix":"987e113518340066","entry":"broadcast_in_dim_equivalent","repo":"deepmind/kfac_jax","repo_kind":"found_in_text","path":"kfac_jax/_src/tag_graph_matcher.py","file_url":"https://github.com/deepmind/kfac_jax/blob/HEAD/kfac_jax/_src/tag_graph_matcher.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"987e113518340066"}},{"code_sha256_prefix":"dc1b986b1be917d4","entry":"compatible_sum","repo":"deepmind/kfac_jax","repo_kind":"found_in_text","path":"kfac_jax/_src/curvature_blocks/utils.py","file_url":"https://github.com/deepmind/kfac_jax/blob/HEAD/kfac_jax/_src/curvature_blocks/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"dc1b986b1be917d4"}},{"code_sha256_prefix":"42474a82097165d1","entry":"eval_jaxpr_eqn","repo":"deepmind/kfac_jax","repo_kind":"found_in_text","path":"kfac_jax/_src/tag_graph_matcher.py","file_url":"https://github.com/deepmind/kfac_jax/blob/HEAD/kfac_jax/_src/tag_graph_matcher.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"42474a82097165d1"}},{"code_sha256_prefix":"3e9f86b459eecc31","entry":"_calc_c_map","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3e9f86b459eecc31"}},{"code_sha256_prefix":"95c482d63878f89f","entry":"_calc_q_slope_1","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"95c482d63878f89f"}},{"code_sha256_prefix":"518f8f86235bcdf2","entry":"_compute_output_params","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"518f8f86235bcdf2"}},{"code_sha256_prefix":"17cfc190249b5d4f","entry":"_estimate_gaussian_mean","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"17cfc190249b5d4f"}},{"code_sha256_prefix":"8bd5c8a0631e8f17","entry":"_leaky_relu","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8bd5c8a0631e8f17"}},{"code_sha256_prefix":"9c3d8825878c6315","entry":"_precompute_or_load_roots","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9c3d8825878c6315"}},{"code_sha256_prefix":"95b55a581b4926c8","entry":"_solve_for_activation_params","repo":"deepmind/dks","repo_kind":"official","path":"dks/base/activation_transform.py","file_url":"https://github.com/deepmind/dks/blob/HEAD/dks/base/activation_transform.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"95b55a581b4926c8"}},{"code_sha256_prefix":"14e10e479569a361","entry":"get_and_verify_loss_meta","repo":"deepmind/kfac_jax","repo_kind":"found_in_text","path":"kfac_jax/_src/layers_and_loss_tags.py","file_url":"https://github.com/deepmind/kfac_jax/blob/HEAD/kfac_jax/_src/layers_and_loss_tags.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"14e10e479569a361"}},{"code_sha256_prefix":"b4a5e9703e6605e1","entry":"get_loss_outputs","repo":"deepmind/kfac_jax","repo_kind":"found_in_text","path":"kfac_jax/_src/layers_and_loss_tags.py","file_url":"https://github.com/deepmind/kfac_jax/blob/HEAD/kfac_jax/_src/layers_and_loss_tags.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b4a5e9703e6605e1"}},{"code_sha256_prefix":"ee339b16cfba7ff0","entry":"reshape_equivalent","repo":"deepmind/kfac_jax","repo_kind":"found_in_text","path":"kfac_jax/_src/tag_graph_matcher.py","file_url":"https://github.com/deepmind/kfac_jax/blob/HEAD/kfac_jax/_src/tag_graph_matcher.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ee339b16cfba7ff0"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}