SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2305.07580 · ICML · 2023

Fisher Information Embedding for Node and Graph Learning

Karsten Borgwardt, Dexiong Chen, Paolo Pellizzoni

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 0 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/2305.07580.json · badge: /api/ran/2305.07580.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
BorgwardtLab/fisher_information_embedding marked official in Papers with Code
Papers with Code archive
3 ran · 0 no recorded run
FunctionStatusWhere it lives
gmm_attn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
BorgwardtLab/fisher_information_embedding/fie/fisher_embeddings.py
code served (permissive licence) · get_code("6b61724f04647160")
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 Tensor, ndim 3, dtype family float, shape (2, 3, 2) 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 torch
import math

torch.manual_seed(0)

# Parameters for gmm_attn function
n = 2  # number of samples
d = 4  # feature dimension
p = 3  # number of mixture components
h = 2  # number of heads

# x: n x d tensor
x = torch.randn(n, d)

# mu: p x h x d tensor
mu = torch.randn(p, h, d)

# log_pi: p x h tensor (bias term)
log_pi = torch.randn(p, h)

# prior: can be None, 'ot', or 'uot'
prior = None

# eps: regularization parameter
eps = 1.0

# gamma: parameter for unbalanced OT
gamma = 1.0

ARGS = {
    'x': x,
    'mu': mu,
    'log_pi': log_pi,
    'prior': prior,
    'eps': eps,
    'gamma': gamma,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 2) — shape (2, 3, 2) 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
ot Ran
ran; our contract draft was wrong, not the code
BorgwardtLab/fisher_information_embedding/fie/fisher_embeddings.py
code served (permissive licence) · get_code("8a516682493fc6e0")
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 Tensor, ndim 2, dtype family float, shape (4, 3) 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 torch
import math

torch.manual_seed(0)

# K is a cost/kernel matrix of shape (m, n)
# The function uses K.shape, K.new_zeros(), K / eps, and indexing operations
# a and b are optional cost vectors or scalars
# eps is a positive float for regularization
# max_iter is a positive int for iterations

K = torch.randn(4, 3)  # m=4, n=3 - distinct dimensions
a = torch.ones(4)      # cost vector for rows (m=4)
b = torch.ones(3)      # cost vector for columns (n=3)
eps = 0.1              # regularization parameter
max_iter = 5           # number of iterations

ARGS = {
    'K': K,
    'a': a,
    'b': b,
    'eps': eps,
    'max_iter': max_iter
}
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 Tensor, ndim 2, dtype family float, shape (4, 3) — shape (4, 3) 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
uot Ran
ran; our contract draft was wrong, not the code
BorgwardtLab/fisher_information_embedding/fie/fisher_embeddings.py
code served (permissive licence) · get_code("b1e86008b3157f09")
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 Tensor, ndim 2, dtype family float, shape (4, 3) 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 torch
import math

torch.manual_seed(0)

# K is a cost/kernel matrix of shape (m, n)
# The function uses K.shape, K.new_zeros(), division, and indexing operations
m, n = 4, 3
K = torch.randn(m, n)

# a is optional, can be None or a tensor
# If not None, it gets torch.log() applied, so needs positive values
a = torch.tensor([1.0, 2.0, 3.0, 0.5])

# b is optional, can be None or a tensor
# If not None, it gets torch.log() applied, so needs positive values
b = torch.tensor([0.5, 1.5, 2.0])

# eps is a scalar float
eps = 1.0

# gamma is a scalar float
gamma = 1.0

# max_iter is an int
max_iter = 5

ARGS = {
    "K": K,
    "a": a,
    "b": b,
    "eps": eps,
    "gamma": gamma,
    "max_iter": max_iter
}
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 Tensor, ndim 2, dtype family float, shape (4, 3) — shape (4, 3) 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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.

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

Attention-based graph neural networks (GNNs), such as graph attention networks (GATs), have become popular neural architectures for processing graph-structured data and learning node embeddings. Despite their empirical success, these models rely on labeled data and the theoretical properties of these models have yet to be fully understood. In this work, we propose a novel attention-based node embedding framework for graphs. Our framework builds upon a hierarchical kernel for multisets of subgraphs around nodes (e.g., neighborhoods) and each kernel leverages the geometry of a smooth statistical manifold to compare pairs of multisets, by "projecting" the multisets onto the manifold. By explicitly computing node embeddings with a manifold of Gaussian mixtures, our method leads to a new attention mechanism for neighborhood aggregation. We provide theoretical insights into generalizability and expressivity of our embeddings, contributing to a deeper understanding of attention-based GNNs. We propose both efficient unsupervised and supervised methods for learning the embeddings. Through experiments on several node classification benchmarks, we demonstrate that our proposed method outperforms existing attention-based graph models like GATs. Our code is available at https://github.com/BorgwardtLab/ fisher_information_embedding.

For agents

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

get_harvested_code_for_paper("2305.07580")
get_code_for_paper("2305.07580")
have("2305.07580")

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

curl https://syntology.ai/api/ran/2305.07580.json

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

[![Syntology run record](https://syntology.ai/api/ran/2305.07580.svg)](https://syntology.ai/paper/2305.07580)

Connect an agent — have() is free.