Methods › General › Distribution Approximation › Normalizing Flows

Normalizing Flows

771 papers tagged archive 2025-07-28

Introduced by Danilo Jimenez Rezende et al. in Variational Inference with Normalizing Flows

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

Normalizing Flows are a method for constructing complex distributions by transforming a probability density through a series of invertible mappings. By repeatedly applying the rule for change of variables, the initial density ‘flows’ through the sequence of invertible mappings. At the end of this sequence we obtain a valid probability distribution and hence this type of flow is referred to as a normalizing flow.

In the case of finite flows, the basic rule for the transformation of densities considers an invertible, smooth mapping f : ℝᵈ →ℝᵈ with inverse f⁻¹ = g, i.e. the composition g ·f(z) = z. If we use this mapping to transform a random variable z with distribution q(z), the resulting random variable z′ = f(z) has a distribution:

q(z′) = q(z)|det(δf⁻¹)/(δz′)|= q(z)|detδf/δz|⁻¹ where the last equality can be seen by applying the chain rule (inverse function theorem) and is a property of Jacobians of invertible functions. We can construct arbitrarily complex densities by composing several simple maps and successively applying the above equation. The density q_K(z) obtained by successively transforming a random variable z₀ with distribution q₀ through a chain of K transformations fₖ is:

z_K = f_K ·…·f₂ ·f₁(z₀)

lnq_K(z_K) = lnq₀(z₀) − ∑ᴷₖ₌₁ln|detδfₖ/(δzₖ₋₁)| The path traversed by the random variables zₖ = fₖ(zₖ₋₁) with initial distribution q₀(z₀) is called the flow and the path formed by the successive distributions qₖ is a normalizing flow.

PaperSourceSee Code · ex4sperans/variational-inference-with-normalizing-flows

In Syntology repository known, no verified implementation (declared_code; repositories: 5 listed of 17 in the graph, strongest link: mentioned on GitHub). This answer is about arXiv:1505.05770, the paper the archive credits with this method.

Code Syntology holds under other names, as the lookup lists them (code_nearby); not a complete list, and none of them is this method.

Origin paper of this answer, arXiv:1505.05770 (its page on this site · its page on Syntology): Syntology's run record for its harvested code, 21 ran (of which 13 constructed an object rather than computing a result; 21 with no instrument failure: 1 honoured, 1 violated, 19 with no contract checked; 0 where Syntology's instrument failed) · 16 unverified (of 37 harvested samples) · 12 pointer-only (licence). The instrument figure counts failures of Syntology's instrument, not of the code. Ran means executed on a synthesized input, not that the code is correct.

Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.

What each level means, quoted from Syntology's server card: V0 “imports”; V1 “executes against a synthesized fixture and returns a value of the recorded type”; V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” The card adds: “No rung on this ladder compares the served code against the authors' implementation, so treat every level as evidence about runnability and internal agreement, not about matching the paper's results.” A level quoted in two parts is one definition the card writes as two clauses joined by a double hyphen, each clause quoted on its own (how these levels are quoted).

Papers archive 2025-07-28

archive papers tagged: 771 · with a code link: 390 · where Syntology ran a sample: 122 (99 with a run with no instrument failure, 23 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (122 of 771 tagged: 99 with a run with no instrument failure, 23 where every run was a failure of Syntology's instrument)

Page 1 of 8: papers 1 to 100 of 771, newest first by the archive's date (ties by slug), in archive order.

Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.

Tasks archive 2025-07-28

20 shown of 367 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Density Estimation76
Variational Inference54
Image Generation37
Anomaly Detection29
Text to Speech26
text-to-speech26
Uncertainty Quantification24
Time Series23
Speech Synthesis22
Bayesian Inference18
Denoising18
regression17
Out-of-Distribution Detection16
Time Series Analysis16
Decision Making15
Gaussian Processes15
Reinforcement Learning (RL)15
Decoder14
Representation Learning14
reinforcement-learning14

Usage over time archive 2025-07-28

Papers per year tagged with Normalizing Flows: 2015 to 2025, peak 139 139 0 2015: 7 papers 2015 2016: 13 papers 2016 2017: 25 papers 2017 2018: 30 papers 2018 2019: 69 papers 2019 2020: 100 papers 2020 2021: 115 papers 2021 2022: 106 papers 2022 2023: 139 papers 2023 2024: 110 papers 2024 2025: 57 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (771 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Distribution Approximation

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections