{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:52FPSCX54XKSAO7XWVIESKS6JR","short_pith_number":"pith:52FPSCX5","schema_version":"1.0","canonical_sha256":"ee8af90afde5d5203bf7b550492a5e4c46bd268faf40cd2a835c37b32cb09d3c","source":{"kind":"arxiv","id":"1912.11554","version":1},"attestation_state":"computed","paper":{"title":"Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"NumPyro composes Pyro effect handlers with JAX to deliver a fully JIT-compiled iterative NUTS sampler.","cross_cats":["cs.AI","cs.LG","cs.PL"],"primary_cat":"stat.ML","authors_text":"Du Phan, Martin Jankowiak, Neeraj Pradhan","submitted_at":"2019-12-24T22:09:36Z","abstract_excerpt":"NumPyro is a lightweight library that provides an alternate NumPy backend to the Pyro probabilistic programming language with the same modeling interface, language primitives and effect handling abstractions. Effect handlers allow Pyro's modeling API to be extended to NumPyro despite its being built atop a fundamentally different JAX-based functional backend. In this work, we demonstrate the power of composing Pyro's effect handlers with the program transformations that enable hardware acceleration, automatic differentiation, and vectorization in JAX. In particular, NumPyro provides an iterati"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"1912.11554","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2019-12-24T22:09:36Z","cross_cats_sorted":["cs.AI","cs.LG","cs.PL"],"title_canon_sha256":"4cada3a002c300a0256c850b4de223b1aecb15768cfd628ce5059fb220f3bb79","abstract_canon_sha256":"23b4291795aec701f8a107f83cc5925ae9b17f75d4488a1480b7e616c2a31232"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:50.993110Z","signature_b64":"Sv0fpw+I8FxFZaKLXISi68PVN6JElAXArY5T1eWTLt24CwTJRmAuYnvu9zIXpdMwaDV6+o6KLV/azWL61ovnBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ee8af90afde5d5203bf7b550492a5e4c46bd268faf40cd2a835c37b32cb09d3c","last_reissued_at":"2026-05-17T23:38:50.992672Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:50.992672Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"NumPyro composes Pyro effect handlers with JAX to deliver a fully JIT-compiled iterative NUTS sampler.","cross_cats":["cs.AI","cs.LG","cs.PL"],"primary_cat":"stat.ML","authors_text":"Du Phan, Martin Jankowiak, Neeraj Pradhan","submitted_at":"2019-12-24T22:09:36Z","abstract_excerpt":"NumPyro is a lightweight library that provides an alternate NumPy backend to the Pyro probabilistic programming language with the same modeling interface, language primitives and effect handling abstractions. Effect handlers allow Pyro's modeling API to be extended to NumPyro despite its being built atop a fundamentally different JAX-based functional backend. In this work, we demonstrate the power of composing Pyro's effect handlers with the program transformations that enable hardware acceleration, automatic differentiation, and vectorization in JAX. In particular, NumPyro provides an iterati"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"NumPyro provides an iterative formulation of the No-U-Turn Sampler (NUTS) that can be end-to-end JIT compiled, yielding an implementation that is much faster than existing alternatives in both the small and large dataset regimes.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That Pyro's effect handlers compose cleanly with JAX's functional transformations without introducing correctness issues or losing expressiveness in the modeling API.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"NumPyro delivers a JIT-compilable iterative NUTS sampler by composing Pyro effect handlers with JAX transformations, achieving faster performance than prior implementations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"NumPyro composes Pyro effect handlers with JAX to deliver a fully JIT-compiled iterative NUTS sampler.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"10f1c0fd0a5e87955ee0e1aae058fed1c814acb0fd873065a02895e3496f2689"},"source":{"id":"1912.11554","kind":"arxiv","version":1},"verdict":{"id":"ec506122-a049-4c3b-ab7e-3c0f7dd3568d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T15:58:48.146930Z","strongest_claim":"NumPyro provides an iterative formulation of the No-U-Turn Sampler (NUTS) that can be end-to-end JIT compiled, yielding an implementation that is much faster than existing alternatives in both the small and large dataset regimes.","one_line_summary":"NumPyro delivers a JIT-compilable iterative NUTS sampler by composing Pyro effect handlers with JAX transformations, achieving faster performance than prior implementations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That Pyro's effect handlers compose cleanly with JAX's functional transformations without introducing correctness issues or losing expressiveness in the modeling API.","pith_extraction_headline":"NumPyro composes Pyro effect handlers with JAX to deliver a fully JIT-compiled iterative NUTS sampler."},"references":{"count":25,"sample":[{"doi":"","year":1912,"title":"Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro","work_id":"a2a03fcf-38c5-4ebe-b5a4-4bc41761dca5","ref_index":1,"cited_arxiv_id":"1912.11554","is_internal_anchor":true},{"doi":"","year":2018,"title":"2018 , copyright =","work_id":"e0d9664f-f519-4c55-b035-0f5e384b0835","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Chen, Martin Jankowiak, Fritz Obermeyer, Neeraj Pradhan, Theofanis Karaletsos, Rohit Singh, Paul Szerlip, Paul Horsfall, and Noah D","work_id":"f70b8d95-49b2-402e-a34c-887bb1b81626","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"Learning disentangled representations with semi-supervised deep generative models","work_id":"d809c362-1be9-4817-bbc6-2a5c9cdc6bfe","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"Automatic differentiation in pytorch","work_id":"5a5818ac-37e1-444f-8802-6256d3fb52b9","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":25,"snapshot_sha256":"2b88815c921d3cad2a3bf203a755ced74764ff24690900947d68ec2b6b187b3a","internal_anchors":3},"formal_canon":{"evidence_count":2,"snapshot_sha256":"7275956bfa88188bfc6605fa18cb416fe8164b8e578a83f3d3ee802222110e4f"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1912.11554","created_at":"2026-05-17T23:38:50.992744+00:00"},{"alias_kind":"arxiv_version","alias_value":"1912.11554v1","created_at":"2026-05-17T23:38:50.992744+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.11554","created_at":"2026-05-17T23:38:50.992744+00:00"},{"alias_kind":"pith_short_12","alias_value":"52FPSCX54XKS","created_at":"2026-05-18T12:33:10.108867+00:00"},{"alias_kind":"pith_short_16","alias_value":"52FPSCX54XKSAO7X","created_at":"2026-05-18T12:33:10.108867+00:00"},{"alias_kind":"pith_short_8","alias_value":"52FPSCX5","created_at":"2026-05-18T12:33:10.108867+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":50,"internal_anchor_count":30,"sample":[{"citing_arxiv_id":"2605.22968","citing_title":"Uncertainty-aware classification and triage of structural heart disease using electrocardiography and echocardiography metrics","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23083","citing_title":"A Strongly Parametrized Mass Ratio Model for the Stable Mass Transfer Channel: a Case Study of the $10 \\, \\rm{M}_{\\odot}$ Peak","ref_index":52,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23134","citing_title":"Archimedean Copula Inference via Taylor-Mode AD","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2503.15821","citing_title":"Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youths with Autism","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2505.09832","citing_title":"LITMUS: Bayesian Lag Recovery in Reverberation Mapping with Fast Differentiable Models","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21615","citing_title":"ASSEMBLAGE-DEEPHISTORY: A Cross-Build Binary Dataset with Temporal Coverage","ref_index":63,"is_internal_anchor":true},{"citing_arxiv_id":"2505.16884","citing_title":"A search for periodic AGN variability in $\\textit{Gaia}$ Data Release 3","ref_index":56,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20345","citing_title":"Corrected Integrated Laplace Approximation for Bayesian Inference in Latent Gaussian Models","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20400","citing_title":"Understanding Deterioration Random Effects for Causal Discovery in Infrastructure Management","ref_index":35,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21253","citing_title":"Theoretical guidelines for annealed Langevin dynamics in compositional simulation-based inference","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20112","citing_title":"Gravitational-wave constraints on $H_0$ are robust to (putative) redshift evolution in the binary black hole mass spectrum at current sensitivity","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18476","citing_title":"AI4BayesCode: From Natural Language Descriptions to Validated Modular Stateful Bayesian Samplers","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2506.03969","citing_title":"Differentiable Fuzzy Cosmic-Web for Field Level Inference","ref_index":77,"is_internal_anchor":true},{"citing_arxiv_id":"2506.09163","citing_title":"Scalable Spatiotemporal Inference with Biased Scan Attention Transformer Neural Processes","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2508.17090","citing_title":"Neural Stochastic Differential Equations on Compact State Spaces: Theory, Methods, and Application to Suicide Risk Modeling","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2508.20266","citing_title":"Using Symbolic Regression to Emulate the Radial Fourier Transform of the S\\'ersic profile for Fast, Accurate and Differentiable Galaxy Profile Fitting","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2509.01082","citing_title":"RefineStat: Efficient Exploration for Probabilistic Program Synthesis","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2509.05219","citing_title":"Conversational AI increases political knowledge as effectively as self-directed internet search","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2510.10924","citing_title":"Image reconstruction with the JWST Interferometer","ref_index":63,"is_internal_anchor":true},{"citing_arxiv_id":"2510.09806","citing_title":"AMIGO: a Data-Driven Calibration of the JWST Interferometer","ref_index":91,"is_internal_anchor":true},{"citing_arxiv_id":"2510.18749","citing_title":"Symbolic Emulators for Cosmology: Accelerating Cosmological Analyses Without Sacrificing Precision","ref_index":56,"is_internal_anchor":true},{"citing_arxiv_id":"2511.09656","citing_title":"Photon counting readout for detection and inference of gravitational waves from neutron star merger remnants","ref_index":59,"is_internal_anchor":true},{"citing_arxiv_id":"2511.15352","citing_title":"People readily follow personal advice from AI but it does not improve their well-being","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2512.01261","citing_title":"OzDES Reverberation Mapping of Active Galactic Nuclei: Final Data Release, Black-Hole Mass Results, & Scaling Relations","ref_index":66,"is_internal_anchor":true},{"citing_arxiv_id":"2512.03229","citing_title":"The DESI DR1 Peculiar Velocity Survey: growth rate measurements from the maximum likelihood fields method","ref_index":101,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR","json":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR.json","graph_json":"https://pith.science/api/pith-number/52FPSCX54XKSAO7XWVIESKS6JR/graph.json","events_json":"https://pith.science/api/pith-number/52FPSCX54XKSAO7XWVIESKS6JR/events.json","paper":"https://pith.science/paper/52FPSCX5"},"agent_actions":{"view_html":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR","download_json":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR.json","view_paper":"https://pith.science/paper/52FPSCX5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1912.11554&json=true","fetch_graph":"https://pith.science/api/pith-number/52FPSCX54XKSAO7XWVIESKS6JR/graph.json","fetch_events":"https://pith.science/api/pith-number/52FPSCX54XKSAO7XWVIESKS6JR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR/action/storage_attestation","attest_author":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR/action/author_attestation","sign_citation":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR/action/citation_signature","submit_replication":"https://pith.science/pith/52FPSCX54XKSAO7XWVIESKS6JR/action/replication_record"}},"created_at":"2026-05-17T23:38:50.992744+00:00","updated_at":"2026-05-17T23:38:50.992744+00:00"}