{"url":"/sota/density-estimation-on-mnist","task":{"name":"Density Estimation","url":"/task/density-estimation","note":null},"dataset":{"name":"MNIST","url":"/dataset/mnist"},"category":"Methodology","categories":["Methodology"],"category_note":null,"description":"The goal of **Density Estimation** is to give an accurate description of the underlying probabilistic density distribution of an observable data set with unknown density.\n\n\n<span class=\"description-source\">Source: [Contrastive Predictive Coding Based Feature for Automatic Speaker Verification ](https://arxiv.org/abs/1904.01575)</span>","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["NLL (bits/dim)","Log-likelihood (nats)","MMD-L2","COV-L2","NLL"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"NLL (bits/dim)":null,"Log-likelihood (nats)":null,"MMD-L2":null,"COV-L2":null,"NLL":null}},"counts":{"rows":6,"rows_with_code":6,"rows_with_paper_page":6,"rows_dated":6,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"Identity","metrics":{"NLL (bits/dim)":"0.134"},"uses_additional_data":false,"paper_date":"2022-05-30","paper":"/paper/gradient-backpropagation-through","paper_url":"https://arxiv.org/abs/2205.15213v3","paper_title":"Backpropagation through Combinatorial Algorithms: Identity with Projection Works","code":"https://github.com/khalil-research/pyepo","n_code_links":2,"syntology":null},{"rank_in_archive_order":2,"model":"RNODE","metrics":{"NLL (bits/dim)":"0.97"},"uses_additional_data":false,"paper_date":"2020-02-07","paper":"/paper/how-to-train-your-neural-ode","paper_url":"https://arxiv.org/abs/2002.02798v3","paper_title":"How to train your neural ODE: the world of Jacobian and kinetic regularization","code":"https://github.com/cfinlay/ffjord-rnode","n_code_links":2,"syntology":{"n_ran":4,"n_unverified":11,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"FFJORD","metrics":{"COV-L2":"96.4%","MMD-L2":"17.3","NLL (bits/dim)":"0.99"},"uses_additional_data":false,"paper_date":"2018-10-02","paper":"/paper/ffjord-free-form-continuous-dynamics-for","paper_url":"http://arxiv.org/abs/1810.01367v3","paper_title":"FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Models","code":"https://github.com/rtqichen/ffjord","n_code_links":7,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"PaddingFlow","metrics":{"COV-L2":"100%","MMD-L2":"11.0"},"uses_additional_data":false,"paper_date":"2024-03-13","paper":"/paper/paddingflow-improving-normalizing-flows-with","paper_url":"https://arxiv.org/abs/2403.08216v2","paper_title":"PaddingFlow: Improving Normalizing Flows with Padding-Dimensional Noise","code":"https://github.com/adamqlmeng/paddingflow","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"DVP-VAE","metrics":{"NLL":"77.1"},"uses_additional_data":false,"paper_date":"2024-12-02","paper":"/paper/hierarchical-vae-with-a-diffusion-based","paper_url":"https://arxiv.org/abs/2412.01373v1","paper_title":"Hierarchical VAE with a Diffusion-based VampPrior","code":"https://github.com/akuzina/dvp_vae","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":6,"model":"MADE MoG","metrics":{"Log-likelihood (nats)":"-1038.5"},"uses_additional_data":false,"paper_date":"2017-05-19","paper":"/paper/masked-autoregressive-flow-for-density","paper_url":"http://arxiv.org/abs/1705.07057v4","paper_title":"Masked Autoregressive Flow for Density Estimation","code":"https://github.com/tensorflow/probability","n_code_links":21,"syntology":{"n_ran":22,"n_unverified":6,"n_samples":28,"n_pointer_only_licence":9}}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":4,"rows_with_any_sample_ran":4,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":31,"n_unverified":19,"n_samples":50,"n_pointer_only_licence":11,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":31,"n_unverified":19,"n_samples":50,"n_pointer_only_licence":11,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}