{"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/som-vae-interpretable-discrete-representation","title":"SOM-VAE: Interpretable Discrete Representation Learning on Time Series","arxiv_id":"1806.02199","date":"2018-06-06","proceeding":"ICLR 2019 5","authors":["Vincent Fortuin","Matthias Hüser","Francesco Locatello","Heiko Strathmann","Gunnar Rätsch"],"abstract":"High-dimensional time series are common in many domains. Since human\ncognition is not optimized to work well in high-dimensional spaces, these areas\ncould benefit from interpretable low-dimensional representations. However, most\nrepresentation learning algorithms for time series data are difficult to\ninterpret. This is due to non-intuitive mappings from data features to salient\nproperties of the representation and non-smoothness over time. To address this\nproblem, we propose a new representation learning framework building on ideas\nfrom interpretable discrete dimensionality reduction and deep generative\nmodeling. This framework allows us to learn discrete representations of time\nseries, which give rise to smooth and interpretable embeddings with superior\nclustering performance. We introduce a new way to overcome the\nnon-differentiability in discrete representation learning and present a\ngradient-based version of the traditional self-organizing map algorithm that is\nmore performant than the original. Furthermore, to allow for a probabilistic\ninterpretation of our method, we integrate a Markov model in the representation\nspace. This model uncovers the temporal transition structure, improves\nclustering performance even further and provides additional explanatory\ninsights as well as a natural representation of uncertainty. We evaluate our\nmodel in terms of clustering performance and interpretability on static\n(Fashion-)MNIST data, a time series of linearly interpolated (Fashion-)MNIST\nimages, a chaotic Lorenz attractor system with two macro states, as well as on\na challenging real world medical time series application on the eICU data set.\nOur learned representations compare favorably with competitor methods and\nfacilitate downstream tasks on the real world data.","url_abs":"http://arxiv.org/abs/1806.02199v7","url_pdf":"http://arxiv.org/pdf/1806.02199v7.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":"som-vae-interpretable-discrete-representation","repo_url":"https://github.com/ratschlab/SOM-VAE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"som-vae-interpretable-discrete-representation","repo_url":"https://github.com/KurochkinAlexey/SOM-VAE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"som-vae-interpretable-discrete-representation","repo_url":"https://github.com/ai-how/TIme-series-clustering","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"som-vae-interpretable-discrete-representation","repo_url":"https://github.com/alexwndm/state-detection-somvae","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"som-vae-interpretable-discrete-representation","repo_url":"https://github.com/shrra/minisom","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"som-vae-interpretable-discrete-representation","repo_url":"https://github.com/merchen911/SOM-VAE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"dimensionality-reduction","task_name":"Dimensionality Reduction"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series","task_name":"Time Series Analysis"},{"task_slug":"time-series-clustering","task_name":"Time Series Clustering"}],"methods":[{"method_slug":"interpretability","method_name":"Interpretability"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/time-series-clustering-on-eicu-collaborative","task":"Time Series Clustering","dataset":"eICU Collaborative Research Database","model":"SOM-VAE-prob","rank_in_archive_order":1,"of":3,"metrics":{"NMI (physiology_24_hours)":"0.0421","NMI (physiology_6_hours)":"0.0474"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-clustering-on-eicu-collaborative","task":"Time Series Clustering","dataset":"eICU Collaborative Research Database","model":"SOM-VAE","rank_in_archive_order":3,"of":3,"metrics":{"NMI (physiology_12_hours)":"0.0444","NMI (physiology_24_hours)":"0.0354","NMI (physiology_6_hours)":"0.0407"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.02199","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1806.02199"}},"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/KurochkinAlexey/SOM-VAE","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ratschlab/SOM-VAE","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/alexwndm/state-detection-somvae","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ai-how/TIme-series-clustering","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/shrra/minisom","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/merchen911/SOM-VAE","reach":{"status":"ok"}}],"summary":{"unverified":6},"by_repo_kind":{"official":{"samples":6,"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":"217082b54273942a","entry":"bias_variable","repo":"ratschlab/SOM-VAE","repo_kind":"official","path":"som_vae/somvae_model.py","file_url":"https://github.com/ratschlab/SOM-VAE/blob/HEAD/som_vae/somvae_model.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":"217082b54273942a"}},{"code_sha256_prefix":"68aa8e556d801b92","entry":"compute_NMI","repo":"ratschlab/SOM-VAE","repo_kind":"official","path":"som_vae/utils.py","file_url":"https://github.com/ratschlab/SOM-VAE/blob/HEAD/som_vae/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":"68aa8e556d801b92"}},{"code_sha256_prefix":"a0a609c5dc52a0e6","entry":"compute_purity","repo":"ratschlab/SOM-VAE","repo_kind":"official","path":"som_vae/utils.py","file_url":"https://github.com/ratschlab/SOM-VAE/blob/HEAD/som_vae/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":"a0a609c5dc52a0e6"}},{"code_sha256_prefix":"ed079f79e2399a16","entry":"conv2d","repo":"ratschlab/SOM-VAE","repo_kind":"official","path":"som_vae/somvae_model.py","file_url":"https://github.com/ratschlab/SOM-VAE/blob/HEAD/som_vae/somvae_model.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":"ed079f79e2399a16"}},{"code_sha256_prefix":"dcccd82a2000d983","entry":"interpolate_arrays","repo":"ratschlab/SOM-VAE","repo_kind":"official","path":"som_vae/utils.py","file_url":"https://github.com/ratschlab/SOM-VAE/blob/HEAD/som_vae/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":"dcccd82a2000d983"}},{"code_sha256_prefix":"27a65c2665dbbd2d","entry":"weight_variable","repo":"ratschlab/SOM-VAE","repo_kind":"official","path":"som_vae/somvae_model.py","file_url":"https://github.com/ratschlab/SOM-VAE/blob/HEAD/som_vae/somvae_model.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":"27a65c2665dbbd2d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}