SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2004.14557 · 2020

Learning Deformable Image Registration from Optimization: Perspective, Modules, Bilevel Training and Beyond

arXiv · PDF · Open in the Atlas

Code that ran

11 of 13 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; 3 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 1 failure of our instrument; 2 no recorded run), measured 2026-11-21; 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/2004.14557.json · badge: /api/ran/2004.14557.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
Alison-brie/MultiPropReg canonical (classifier label)
classifier basis: census
11 ran · 2 no recorded run
FunctionStatusWhere it lives
MINDSSC Ran
ran on a synthesized input
Alison-brie/MultiPropReg/MIND.py
code served (permissive licence) · get_code("1fd1a81503e31013")
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 5, dtype family float, shape (2, 12, 8, 8, 8) as a later run recorded it. 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, 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 torch.nn as nn

torch.manual_seed(0)

# Create a small 3D image tensor
# The function expects a 5D tensor (batch, channels, depth, height, width)
# Based on the code: img is used with F.conv3d and avg_pool3d
# kernel_size will be radius*2+1 = 5 with default radius=2
# We need spatial dims large enough for pooling operations
img = torch.randn(2, 1, 8, 8, 8)

ARGS = {
    "img": img,
    "radius": 2,
    "dilation": 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 Tensor, ndim 5, dtype family float, shape (2, 12, 8, 8, 8) — shape (2, 12, 8, 8, 8) 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
boundingbox Ran
ran, honoured the contract we drafted
Alison-brie/MultiPropReg/ext/pynd/ndutils.py
code served (permissive licence) · get_code("c516c53184ea0a6b")
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 ndarray, ndim 1, dtype family int, shape (6,) 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

# Create a small binary volume (nd array)
# The function finds bounding box of non-zero elements
# Using a 3D array as an example
bwvol = np.zeros((5, 6, 7), dtype=np.uint8)
# Add some non-zero values to define a bounding box
bwvol[1:3, 2:4, 3:5] = 1

ARGS = {
    "bwvol": bwvol
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family int, shape (6,) — shape (6,) 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
bw2sdtrf Ran
ran, honoured the contract we drafted
Alison-brie/MultiPropReg/ext/pynd/ndutils.py
code served (permissive licence) · get_code("52abec54f9082f54")
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 ndarray, 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 numpy as np
import scipy.ndimage

# Create a small binary volume for testing
# Using a 4x3x2 volume with some True values and some False values
bwvol = np.array([
    [[True, False],
     [False, True],
     [True, True]],
    [[False, True],
     [True, False],
     [False, False]]
])

ARGS = {
    "bwvol": bwvol
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, 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
bwdist Ran
ran, honoured the contract we drafted
Alison-brie/MultiPropReg/ext/pynd/ndutils.py
code served (permissive licence) · get_code("7d3d4ac668b84a80")
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 ndarray, ndim 3, dtype family float, shape (4, 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 numpy as np
import scipy.ndimage

# Create a small logical volume (binary array) for testing
# Using a 4x3x2 array to have distinct dimensions
bwvol = np.array([
    [[True, False], [False, True], [True, False]],
    [[False, True], [True, False], [False, True]],
    [[True, True], [False, False], [True, True]],
    [[False, False], [True, True], [False, False]]
], dtype=bool)

ARGS = {
    "bwvol": bwvol
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (4, 3, 2) — shape (4, 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
compute_local_sums Ran
ran on a synthesized input
Alison-brie/MultiPropReg/losses.py
code served (permissive licence) · get_code("0587a796d489669a")
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 tuple, ndim 6, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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
import numpy as np

torch.manual_seed(0)

# Create small 3D tensors for I and J
# Shape: (batch=2, channels=1, depth=4, height=3, width=3)
I = torch.randn(2, 1, 4, 3, 3)
J = torch.randn(2, 1, 4, 3, 3)

# Create a filter for conv3d
# Shape: (out_channels=1, in_channels=1, kernel_d=2, kernel_h=2, kernel_w=2)
win = (2, 2, 2)
filt = torch.ones(1, 1, 2, 2, 2)

# Parameters for conv3d
stride = 1
padding = 0

ARGS = {
    "I": I,
    "J": J,
    "filt": filt,
    "stride": stride,
    "padding": padding,
    "win": win
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 6, dtype family float — a tuple output has no array shape of its own; none is stated
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
conv Ran
ran on a synthesized input
Alison-brie/MultiPropReg/MultiModal/layers.py
code served (permissive licence) · get_code("44191cc43b143bfd")
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 Sequential. 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

torch.manual_seed(0)

ARGS = {
    'in_planes': 3,
    'out_planes': 16,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'dilation': 1,
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
dice Ran
ran on a synthesized input
Alison-brie/MultiPropReg/utils.py
code served (permissive licence) · get_code("67c81b2eb8a2cf43")
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 ndarray, ndim 1, dtype family float, shape (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 numpy as np

# Create two small volume arrays with labels
vol1 = np.array([
    [[0, 1, 1, 2],
     [0, 1, 2, 2],
     [3, 3, 0, 0],
     [3, 0, 0, 0]],
    [[0, 1, 1, 2],
     [0, 1, 2, 2],
     [3, 3, 0, 0],
     [3, 0, 0, 0]]
])

vol2 = np.array([
    [[0, 1, 1, 0],
     [0, 1, 2, 2],
     [3, 3, 0, 0],
     [0, 0, 0, 0]],
    [[0, 1, 0, 2],
     [0, 1, 2, 2],
     [3, 0, 0, 0],
     [3, 0, 0, 0]]
])

ARGS = {
    'vol1': vol1,
    'vol2': vol2,
    'labels': None,
    'nargout': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (3,) — shape (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
mind_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Alison-brie/MultiPropReg/MIND.py
code served (permissive licence) · get_code("bc93fe252379534f")
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 0, dtype family float, shape () as a later run recorded it. 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, 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 torch.nn as nn

torch.manual_seed(0)

# The function mind_loss takes two arguments: x and y
# Both are passed to MINDSSC, which expects 3D images
# MINDSSC performs 3D convolutions, so input shape should be (batch, channels, depth, height, width)
# We'll use tiny dimensions: batch=2, channels=1, depth=4, height=3, width=3

x = torch.randn(2, 1, 4, 3, 3)
y = torch.randn(2, 1, 4, 3, 3)

ARGS = {
    'x': x,
    'y': y,
}
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 0, dtype family float, shape () — shape () 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
ncc Ran
ran on a synthesized input
Alison-brie/MultiPropReg/losses.py
code served (permissive licence) · get_code("7d0ffd3245e38bd5")
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 0, dtype family float, shape () as a later run recorded it. 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, 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 torch.nn.functional as F
import numpy as np
import math

torch.manual_seed(0)

# The function ncc takes two parameters: I and J
# Both should be tensors that work with F.conv3d
# F.conv3d expects input of shape (batch, channels, depth, height, width)
# The function checks ndims = len(I.size()) - 2, so for 3D it expects 5D tensor

# Create small 3D tensors (batch=2, channels=1, depth=4, height=3, width=5)
I = torch.randn(2, 1, 4, 3, 5, dtype=torch.float32)
J = torch.randn(2, 1, 4, 3, 5, dtype=torch.float32)

ARGS = {
    "I": I,
    "J": J,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () 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
pdist_squared Ran
ran on a synthesized input
Alison-brie/MultiPropReg/MIND.py
code served (permissive licence) · get_code("ebd04f809925dd6e")
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 3, dtype family float, shape (2, 3, 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

torch.manual_seed(0)

# The function pdist_squared computes pairwise squared distances
# It expects x to be a 3D tensor with shape (batch_size, num_features, num_points)
# or (batch_size, num_points, num_features) depending on the operation
# Looking at the code: x ** 2 sums over dim=1, then unsqueeze(2) adds a dimension
# x.permute(0, 2, 1) is used for bmm, suggesting x is (batch, features, points)

# Create a small batch of feature vectors
# Shape: (batch_size=2, num_features=4, num_points=3)
x = torch.randn(2, 4, 3)

ARGS = {
    "x": x
}
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 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 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
predict_flow Ran
ran on a synthesized input
Alison-brie/MultiPropReg/MultiModal/layers.py
code served (permissive licence) · get_code("06221ade5bbdb0bf")
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 Conv3d. 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

torch.manual_seed(0)

ARGS = {
    "in_planes": 64
}
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 Conv3d — a Conv3d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv3d 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
MatchCost Not yet run
unverified (harvested, no recorded run)
Alison-brie/MultiPropReg/MultiModal/layers.py
code served (permissive licence) · get_code("3991b1d1d129c001")
How it ran

no invocation recorded.

count_parameters_in_MB Not yet run
unverified (harvested, no recorded run)
Alison-brie/MultiPropReg/utils.py
code served (permissive licence) · get_code("3e816289579cc9e9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 5 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 4 later) and 8 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
count_parameters_in_MB 8 other papers
2 dated earlier, 6 later
1806.09055 (2018-06)
DARTS: Differentiable Architecture Search
1912.00195 (2019-12)
SGAS: Sequential Greedy Architecture Search
2006.16537 (2020-06)
Theory-Inspired Path-Regularized Differential Network Architecture Search
and 5 more not named here
boundingbox 1 other paper
1 later
2105.04349 (2021-05)
Generative Adversarial Registration for Improved Conditional Deformable Templates
bw2sdtrf 1 other paper
1 later
2105.04349 (2021-05)
Generative Adversarial Registration for Improved Conditional Deformable Templates
bwdist 1 other paper
1 later
2105.04349 (2021-05)
Generative Adversarial Registration for Improved Conditional Deformable Templates
mind_loss 1 other paper
1 dated earlier
1911.12408 (2019-11)
PointPWC-Net: A Coarse-to-Fine Network for Supervised and Self-Supervised Scene Flow Estimation on 3D Point Clouds

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Conventional deformable registration methods aim at solving an optimization model carefully designed on image pairs and their computational costs are exceptionally high. In contrast, recent deep learning based approaches can provide fast deformation estimation. These heuristic network architectures are fully data-driven and thus lack explicit geometric constraints, e.g., topology-preserving, which are indispensable to generate plausible deformations. We design a new deep learning based framework to optimize a diffeomorphic model via multi-scale propagation in order to integrate advantages and avoid limitations of these two categories of approaches. Specifically, we introduce a generic optimization model to formulate diffeomorphic registration and develop a series of learnable architectures to obtain propagative updating in the coarse-to-fine feature space. Moreover, we propose a novel bilevel self-tuned training strategy, allowing efficient search of task-specific hyper-parameters. This training strategy increases the flexibility to various types of data while reduces computational and human burdens. We conduct two groups of image registration experiments on 3D volume datasets including image-to-atlas registration on brain MRI data and image-to-image registration on liver CT data. Extensive results demonstrate the state-of-the-art performance of the proposed method with diffeomorphic guarantee and extreme efficiency. We also apply our framework to challenging multi-modal image registration, and investigate how our registration to support the down-streaming tasks for medical image analysis including multi-modal fusion and image segmentation.

For agents

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

get_harvested_code_for_paper("2004.14557")
get_code_for_paper("2004.14557")
have("2004.14557")

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

curl https://syntology.ai/api/ran/2004.14557.json

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

[![Syntology run record](https://syntology.ai/api/ran/2004.14557.svg)](https://syntology.ai/paper/2004.14557)

Connect an agent — have() is free.