Papers › Gradient Boosted Normalizing Flows

Gradient Boosted Normalizing Flows

27 Feb 2020NeurIPS 2020 12arXiv:2002.11896archive 2025-07-28

Robert Giaquinto, Arindam Banerjee

By chaining a sequence of differentiable invertible transformations, normalizing flows (NF) provide an expressive method of posterior approximation, exact density evaluation, and sampling. The trend in normalizing flow literature has been to devise deeper, more complex transformations to achieve greater flexibility. We propose an alternative: Gradient Boosted Normalizing Flows (GBNF) model a density by successively adding new NF components with gradient boosting. Under the boosting framework, each new NF component optimizes a sample weighted likelihood objective, resulting in new components that are fit to the residuals of the previously trained components. The GBNF formulation results in a mixture model structure, whose flexibility increases as more components are added. Moreover, GBNFs offer a wider, as opposed to strictly deeper, approach that improves existing NFs at the cost of additional training---not more complex transformations. We demonstrate the effectiveness of this technique for density estimation and, by coupling GBNF with a variational autoencoder, generative modeling of images. Our results show that GBNFs outperform their non-boosted analog, and, in some cases, produce better results with smaller, simpler flows.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 30 code samples harvested from 1 repository linked to this paper: 20 ran (of which 0 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 3 where Syntology's instrument failed) · 10 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 30 samples from 1 repository, 20 ran (17 with no instrument failure; 3 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

robert-giaquinto/gradient-boosted-normalizing-flows officialmentioned on GitHubpytorchSyntology: 30 samples, 20 ran (3 instrument-failed) · 10 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

30 samples harvested: 20 ran (of which 0 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 3 where Syntology's instrument failed) · 10 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · our draft was wrong
17ran
10unverified

Licence: 0 of the 30 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from robert-giaquinto/gradient-boosted-normalizing-flows. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: 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. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

robert-giaquinto/gradient-boosted-normalizing-flows official repository 30 samples, 20 ran (3 instrument-failed) · 10 unverified
ActNorm1d models/boosted_flow.py ran MIT (permissive) · 7315468727393ef4 · report fixture: returned tuple · level recorded 2026-09-24
ActNorm2d models/boosted_flow.py ran · metamorphic tier: deterministic MIT (permissive) · 5dfed0f3161fd4b1 · report fixture: returned tuple · level recorded 2026-09-22
BatchNorm in 19 papers models/boosted_flow.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 84ebb2409e0a1cc7 · report fixture: signature x:Tensor/2/float · returned tuple · level recorded 2026-09-22
Conv2d in 43 papers
census: 7 implementations from 7 papers, 6 distinct outputs on one shared input
models/boosted_flow.py ran MIT (permissive) · 2d7d6fd92e4989a5 · report fixture: returned Tensor · level recorded 2026-09-24
Conv2dZeros models/boosted_flow.py ran · metamorphic tier: invariant MIT (permissive) · eeed4e16d893cea5 · report fixture: signature sample:Tensor/4/float · returned Tensor · level recorded 2026-09-22
InvertibleConv1x1 models/boosted_flow.py ran · metamorphic tier: deterministic MIT (permissive) · 5c8a917425f0def4 · report fixture: signature sample:Tensor/4/float, logdet:Tensor/0/float, reverse:bool/0/bool · returned tuple · level recorded 2026-09-22
LinearZeros models/boosted_flow.py ran · metamorphic tier: invariant MIT (permissive) · 11e0e1eaee63a699 · report fixture: signature sample:Tensor/2/float · returned Tensor · level recorded 2026-09-22
Permute1d models/boosted_flow.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · ec45833a8bb10b3c · report fixture: signature sample:Tensor/2/float, reverse:bool/0/bool · returned Tensor · level recorded 2026-09-22
Permute2d models/boosted_flow.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 3884e401a1dea604 · report fixture: signature sample:Tensor/4/float, reverse:bool/0/bool · returned Tensor · level recorded 2026-09-22
PermuteNd models/boosted_flow.py ran MIT (permissive) · d6de36917b79ca67 · report fixture: signature forward:Tensor/4/float, reverse:bool/0/bool · returned NoneType · level recorded 2026-09-24
ReLUNet models/boosted_flow.py ran · metamorphic tier: invariant MIT (permissive) · b27b3f12e9fea1e8 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
RealNVP in 5 papers models/boosted_flow.py ran · metamorphic tier: deterministic MIT (permissive) · 55233dad2a63c90e · report fixture: signature x:Tensor/2/float, layers:tuple/1/tuple · returned tuple · level recorded 2026-09-22
ResidualBlock in 100 papers models/boosted_flow.py ran · metamorphic tier: invariant MIT (permissive) · 0a7417c78d382ffd · report fixture: signature inputs:Tensor/2/float · returned Tensor · level recorded 2026-09-22
ResidualNet models/boosted_flow.py ran · metamorphic tier: invariant MIT (permissive) · 28dd15935b6ad2c2 · report fixture: signature inputs:Tensor/2/float · returned Tensor · level recorded 2026-09-22
Split2d models/boosted_flow.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · b46eb8d0c2c169c9 · report fixture: signature sample:Tensor/4/float, logdet:float/0/float, reverse:bool/0/bool, temperature:float/0/float · returned tuple · level recorded 2026-09-22
SqueezeLayer models/boosted_flow.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 825e0d0a10f8281b · report fixture: signature sample:Tensor/4/float, logdet:NoneType/0/?, reverse:bool/0/bool · returned tuple · level recorded 2026-09-22
TanhNet models/boosted_flow.py ran · metamorphic tier: invariant MIT (permissive) · 45e491a0b611d3dc · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
log_normal_diag in 5 papers models/boosted_flow.py ran · our draft was wrong fingerprinted MIT (permissive) · de88c452ca85eb2d · report fixture: signature x:Tensor/2/float, mean:Tensor/2/float, log_var:Tensor/2/float, average:bool/0/bool, reduce:bool/0/bool, dim:NoneType/0/? · returned Tensor · level recorded 2026-09-22
log_normal_standard models/boosted_flow.py ran · our draft was wrong fingerprinted MIT (permissive) · 6ee1f71f50ee4b26 · report fixture: signature x:Tensor/2/float, average:bool/0/bool, reduce:bool/0/bool, dim:NoneType/0/?, device:NoneType/0/? · returned Tensor · level recorded 2026-09-22
split_feature in 8 papers models/boosted_flow.py ran · our draft was wrong MIT (permissive) · 45f3824dcc7faec1 · report fixture: signature tensor:Tensor/4/float, type:str/0/str · returned tuple · level recorded 2026-09-22
BoostedFlow models/boosted_flow.py unverified MIT (permissive) · 639270ba7c0b6ccd · report
ConvNet in 13 papers models/boosted_flow.py unverified MIT (permissive) · 85e265af1f8b4721 · report
FlowNet models/boosted_flow.py unverified MIT (permissive) · e898fc9ab950c09b · report
FlowStep models/boosted_flow.py unverified MIT (permissive) · cfe3b31cd1ff04b1 · report
GenerativeFlow models/boosted_flow.py unverified MIT (permissive) · 3b26d29c4a8c9004 · report
Glow models/boosted_flow.py unverified MIT (permissive) · 832c9c612efadf21 · report
RealNVPFlow models/boosted_flow.py unverified MIT (permissive) · bd9b9fc7030ba799 · report
_ActNorm models/boosted_flow.py unverified MIT (permissive) · 0e3e13be358d4e88 · report
_Dequantization models/boosted_flow.py unverified MIT (permissive) · 1ec304d03cf1726b · report
epoch_reporting density_experiment.py unverified MIT (permissive) · ebb3df3128ed7f4a · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for Gradient Boosted Normalizing Flows (graph method: Gradient Boosted Normalizing Flows). 33 bricks; individually on a fixture: 24 ran (of which 0 constructed an object rather than computing a result; 19 with no instrument failure: 2 honoured, 0 violated, 17 with no contract checked; 5 where Syntology's instrument failed) · 9 unverified. The assembly has not been run end-to-end.

  1. BoostedFlow entry · level 0 · unverified · 639270ba7c0b6ccd
  2. GenerativeFlow helper · level 0 · unverified · 3b26d29c4a8c9004
  3. Glow helper · level 0 · unverified · 832c9c612efadf21
  4. RealNVPFlow helper · level 0 · unverified · bd9b9fc7030ba799
  5. log_normal_standard helper · level 1 · ran · our draft was wrong · 6ee1f71f50ee4b26
  6. ActNorm2d helper · level 1 · ran · 5dfed0f3161fd4b1
  7. Conv2dZeros helper · level 1 · ran · eeed4e16d893cea5
  8. FlowNet helper · level 0 · unverified · e898fc9ab950c09b
  9. LinearZeros helper · level 1 · ran · 11e0e1eaee63a699
  10. _Dequantization helper · level 0 · unverified · 1ec304d03cf1726b
  11. split_feature helper · level 1 · ran · our draft was wrong · 45f3824dcc7faec1
  12. BatchNorm helper · level 1 · ran · 84ebb2409e0a1cc7
  13. ReLUNet helper · level 1 · ran · b27b3f12e9fea1e8
  14. RealNVP model · level 1 · ran · 55233dad2a63c90e
  15. ResidualNet helper · level 1 · ran · 28dd15935b6ad2c2
  16. TanhNet helper · level 1 · ran · 45e491a0b611d3dc
  17. _ActNorm helper · level 0 · unverified · 0e3e13be358d4e88
  18. compute_same_pad helper · level 1 · ran · honoured contract · 1493e9433f206e03
  19. FlowStep helper · level 0 · unverified · cfe3b31cd1ff04b1
  20. Split2d helper · level 1 · ran · b46eb8d0c2c169c9
  21. SqueezeLayer helper · level 1 · ran · 825e0d0a10f8281b
  22. safe_log helper · level 1 · ran · honoured contract · 6298a45a525420e8
  23. ResidualBlock helper · level 1 · ran · 0a7417c78d382ffd
  24. ActNorm1d helper · level 1 · ran · 7315468727393ef4
  25. ConvNet helper · level 0 · unverified · 85e265af1f8b4721
  26. InvertibleConv1x1 helper · level 1 · ran · 5c8a917425f0def4
  27. Permute1d helper · level 1 · ran · ec45833a8bb10b3c
  28. Permute2d helper · level 1 · ran · 3884e401a1dea604
  29. log_normal_diag helper · level 1 · ran · our draft was wrong · de88c452ca85eb2d
  30. squeeze2d helper · level 1 · ran; the contract check could not run · b503e88f973c2713
  31. unsqueeze2d helper · level 1 · ran; the contract check could not run · 3fa058c6e36e8e99
  32. Conv2d helper · level 1 · ran · 2d7d6fd92e4989a5
  33. PermuteNd helper · level 1 · ran · d6de36917b79ca67

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2002.11896") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Density EstimationVariational Inference

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Normalizing FlowsStochastic Gradient Variational Bayes

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