{"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/bayesgrad-explaining-predictions-of-graph","title":"BayesGrad: Explaining Predictions of Graph Convolutional Networks","arxiv_id":"1807.01985","date":"2018-07-04","proceeding":null,"authors":["Hirotaka Akita","Kosuke Nakago","Tomoki Komatsu","Yohei Sugawara","Shin-ichi Maeda","Yukino Baba","Hisashi Kashima"],"abstract":"Recent advances in graph convolutional networks have significantly improved\nthe performance of chemical predictions, raising a new research question: \"how\ndo we explain the predictions of graph convolutional networks?\" A possible\napproach to answer this question is to visualize evidence substructures\nresponsible for the predictions. For chemical property prediction tasks, the\nsample size of the training data is often small and/or a label imbalance\nproblem occurs, where a few samples belong to a single class and the majority\nof samples belong to the other classes. This can lead to uncertainty related to\nthe learned parameters of the machine learning model. To address this\nuncertainty, we propose BayesGrad, utilizing the Bayesian predictive\ndistribution, to define the importance of each node in an input graph, which is\ncomputed efficiently using the dropout technique. We demonstrate that BayesGrad\nsuccessfully visualizes the substructures responsible for the label prediction\nin the artificial experiment, even when the sample size is small. Furthermore,\nwe use a real dataset to evaluate the effectiveness of the visualization. The\nbasic idea of BayesGrad is not limited to graph-structured data and can be\napplied to other data types.","url_abs":"http://arxiv.org/abs/1807.01985v1","url_pdf":"http://arxiv.org/pdf/1807.01985v1.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":"bayesgrad-explaining-predictions-of-graph","repo_url":"https://github.com/pfnet-research/bayesgrad","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"property-prediction","task_name":"Property Prediction"}],"methods":[{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"graph-convolutional-networks","method_name":"Graph Convolutional Networks"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.01985","atlas_url":"https://app.syntology.ai/?focus=1807.01985","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1807.01985"}},"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/pfnet-research/bayesgrad","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":4},"by_repo_kind":{"official":{"samples":4,"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":"7b427b7a292aee02","entry":"calc_recall_precision","repo":"pfnet-research/bayesgrad","repo_kind":"official","path":"experiments/tox21/plot_precision_recall.py","file_url":"https://github.com/pfnet-research/bayesgrad/blob/HEAD/experiments/tox21/plot_precision_recall.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":"7b427b7a292aee02"}},{"code_sha256_prefix":"1795f8f024d326f7","entry":"calc_recall_precision_for_rate","repo":"pfnet-research/bayesgrad","repo_kind":"official","path":"experiments/tox21/plot_precision_recall.py","file_url":"https://github.com/pfnet-research/bayesgrad/blob/HEAD/experiments/tox21/plot_precision_recall.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":"1795f8f024d326f7"}},{"code_sha256_prefix":"faff5ff0e5fabfb8","entry":"load_npz","repo":"pfnet-research/bayesgrad","repo_kind":"official","path":"experiments/tox21/utils.py","file_url":"https://github.com/pfnet-research/bayesgrad/blob/HEAD/experiments/tox21/utils.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":"faff5ff0e5fabfb8"}},{"code_sha256_prefix":"38cf2b278a31129f","entry":"percentile_index","repo":"pfnet-research/bayesgrad","repo_kind":"official","path":"experiments/tox21/plot_precision_recall.py","file_url":"https://github.com/pfnet-research/bayesgrad/blob/HEAD/experiments/tox21/plot_precision_recall.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":"38cf2b278a31129f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}