SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2201.12245 · NeurIPS · 2022

Wasserstein Iterative Networks for Barycenter Estimation

Alexander Korotin, Vage Egiazarian, Lingxiao Li, Evgeny Burnaev

arXiv · PDF · Open in the Atlas

Code that ran

3 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 1 failure of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2201.12245.json · badge: /api/ran/2201.12245.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
iamalexkorotin/wassersteiniterativenetworks canonical (classifier label)
classifier basis: classifier_2025
3 ran · 2 no recorded run
FunctionStatusWhere it lives
ConvexQuadratic Ran
ran on a synthesized input · constructed an object; no result was computed
iamalexkorotin/wassersteiniterativenetworks/src/icnn.py
code served (permissive licence) · get_code("940bc5a66747a0dc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, ConvexQuadratic: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import torch.nn.functional as F

torch.manual_seed(0)

class ConvexQuadratic(nn.Module):
    """Convex Quadratic Layer"""
    __constants__ = ['in_features', 'out_features', 'quadratic_decomposed', 'weight', 'bias']

    def __init__(self, in_features, out_features, bias=True, rank=1):
        super(ConvexQuadratic, self).__init__()
        self.in_features = in_features
        self.out_features = out_features
        self.rank = rank
        self.quadratic_decomposed = nn.Parameter(torch.Tensor(torch.randn(in_features, rank, out_features)))
        self.weight = nn.Parameter(torch.Tensor(torch.randn(out_features, in_features)))
        if bias:
            self.bias = nn.Parameter(torch.Tensor(out_features))
        else:
            self.register_parameter('bias', None)

    def forward(self, input):
        quad = (input.matmul(self.quadratic_decomposed.transpose(1, 0)).transpose(1, 0) ** 2).sum(dim=1)
        linear = F.linear(input, self.weight, self.bias)
        return quad + linear

# Create a test input tensor (batch_size=2, in_features=4)
test_input = torch.randn(2, 4)

ARGS = {
    'in_features': 4,
    'out_features': 3,
    'bias': True,
    'rank': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ConvexQuadratic — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ConvexQuadratic output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
DenseICNN_U Ran
ran on a synthesized input · constructed an object; no result was computed
iamalexkorotin/wassersteiniterativenetworks/src/icnn.py
code served (permissive licence) · get_code("ebc519711d783219")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, DenseICNN_U: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'in_dim': 4,
    'hidden_layer_sizes': [8, 16, 32],
    'rank': 2,
    'activation': 'celu',
    'dropout': 0.03
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DenseICNN_U — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a DenseICNN_U output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_barycenter_cov Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
iamalexkorotin/wassersteiniterativenetworks/src/bar_benchmark.py
code served (permissive licence) · get_code("645282f7351e83ea")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned ndarray, ndim 2, dtype family float, shape (4, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from scipy.stats import ortho_group

# Set random seed for reproducibility
np.random.seed(0)

# Create small covariance matrices
# Covariance matrices must be symmetric positive definite
dim = 4
num_covs = 2

covs = []
for i in range(num_covs):
    # Generate random orthogonal matrix
    Q = ortho_group.rvs(dim)
    # Generate random positive eigenvalues
    eigenvalues = np.array([1.0 + i * 0.5 for i in range(dim)], dtype=np.float32)
    # Construct symmetric positive definite matrix: Q @ diag(eigenvalues) @ Q.T
    cov = Q @ np.diag(eigenvalues) @ Q.T
    cov = np.real(cov).astype(np.float32)
    covs.append(cov)

# Create weights that sum to 1
alphas = np.array([0.3, 0.7], dtype=np.float32)

# Define arguments
ARGS = {
    "covs": covs,
    "alphas": alphas,
    "max_iter": 10,
    "tol": 1e-8,
    "verbose": False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (4, 4) — shape (4, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_linear_transport Not yet run
unverified (harvested, no recorded run)
iamalexkorotin/wassersteiniterativenetworks/src/bar_benchmark.py
code served (permissive licence) · get_code("47daf6e7349ed8c6")
How it ran

no invocation recorded. this sample carries JSON test cases instead of a setup block; get_reference_implementation serves them with the level they back.

symmetrize Not yet run
unverified (harvested, no recorded run)
iamalexkorotin/wassersteiniterativenetworks/src/bar_benchmark.py
code served (permissive licence) · get_code("06fde2e68ab125fa")
How it ran

no invocation recorded. this sample carries JSON test cases instead of a setup block; get_reference_implementation serves them with the level they back.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 1 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
symmetrize 2 other papers
1 dated earlier, 1 later
2106.01954 (2021-06)
Do Neural Optimal Transport Solvers Work? A Continuous Wasserstein-2 Benchmark
2306.10161 (2023-06)
Building the Bridge of Schrödinger: A Continuous Entropic Optimal Transport Benchmark
get_barycenter_cov 1 other paper
1 dated earlier
2106.01954 (2021-06)
Do Neural Optimal Transport Solvers Work? A Continuous Wasserstein-2 Benchmark
get_linear_transport 1 other paper
1 dated earlier
2106.01954 (2021-06)
Do Neural Optimal Transport Solvers Work? A Continuous Wasserstein-2 Benchmark

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Wasserstein barycenters have become popular due to their ability to represent the average of probability measures in a geometrically meaningful way. In this paper, we present an algorithm to approximate the Wasserstein-2 barycenters of continuous measures via a generative model. Previous approaches rely on regularization (entropic/quadratic) which introduces bias or on input convex neural networks which are not expressive enough for large-scale tasks. In contrast, our algorithm does not introduce bias and allows using arbitrary neural networks. In addition, based on the celebrity faces dataset, we construct Ave, celeba! dataset which can be used for quantitative evaluation of barycenter algorithms by using standard metrics of generative models such as FID. 36th Conference on Neural Information Processing Systems (NeurIPS 2022).

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2201.12245")
get_code_for_paper("2201.12245")
have("2201.12245")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2201.12245.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2201.12245.svg)](https://syntology.ai/paper/2201.12245)

Connect an agent — have() is free.