SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.12373 · NeurIPS · 2022

Learning Proper Orthogonal Decomposition of Complex Dynamics Using Heavy-ball Neural ODEs

Bao Wang, Justin Baker, Elena Cherkaev, Akil Narayan

arXiv · PDF · Open in the Atlas

Code that ran

12 of 15 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; 1 honoured the contract we drafted, 0 violated it and 10 have no contract verdict; 1 failure of our instrument; 3 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/2202.12373.json · badge: /api/ran/2202.12373.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
justinbakermath/pod_hbnode canonical (classifier label)
classifier basis not recorded on the link
12 ran · 3 no recorded run
FunctionStatusWhere it lives
DMD Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/dmd.py
code served (permissive licence) · get_code("e090df891e297576")
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. 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 numpy as np

np.random.seed(0)

# DMD expects:
# X: matrix of shape (n_features, n_snapshots)
# Xp: matrix of shape (n_features, n_snapshots) - typically the "next" snapshots
# modes: integer - number of modes to use in SVD truncation

n_features = 5
n_snapshots = 4
modes = 3

# Create X as a random matrix
X = np.random.randn(n_features, n_snapshots)

# Create Xp as a random matrix (same shape as X)
Xp = np.random.randn(n_features, n_snapshots)

ARGS = {
    'X': X,
    'Xp': Xp,
    'modes': modes
}
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 tuple — a tuple output has no array shape of its own; none is 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
DMD1 Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/dmd.py
code served (permissive licence) · get_code("39e053a8fc66694c")
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. 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 numpy as np

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

# Create sample data
# data should be a 2D numpy array with shape (time_steps, features)
# Using small dimensions: 10 time steps, 4 features
data = np.random.randn(10, 4)

# Define parameters for DMD1
s_ind = 0          # start index
e_ind = 10         # end index
modes = 2          # number of modes to extract
lifts = ()         # empty tuple for lifts parameter

ARGS = {
    "data": data,
    "s_ind": s_ind,
    "e_ind": e_ind,
    "modes": modes,
    "lifts": lifts
}
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 tuple — a tuple output has no array shape of its own; none is 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
DMD2 Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/dmd.py
code served (permissive licence) · get_code("70bae7e4da0f56e1")
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. 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 numpy as np

# Create small synthetic data matching the expected shape
# data should have shape (time_steps, spatial_x, spatial_y, variables)
# We'll use: time=5, spatial_x=3, spatial_y=2, variables=2

np.random.seed(0)

time_steps = 5
spatial_x = 3
spatial_y = 2
num_vars = 2

# Create synthetic data with shape (time, spatial_x, spatial_y, num_vars)
data = np.random.randn(time_steps, spatial_x, spatial_y, num_vars)

# Define arguments for DMD2 function
ARGS = {
    'data': data,           # shape: (5, 3, 2, 2)
    's_ind': 0,             # start index
    'e_ind': 4,             # end index (exclusive, so uses indices 0-3 = 4 timesteps)
    'modes': 2,             # number of DMD modes to extract
    'lifts': ()             # no lifting functions applied
}
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 tuple — a tuple output has no array shape of its own; none is 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
EE_DAT Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/network.py
code served (permissive licence) · get_code("b35ab2aa835ed3b1")
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, 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 numpy as np
import tempfile
import os

# Create a temporary npz file with the required arrays
temp_dir = tempfile.gettempdir()
npz_file = os.path.join(temp_dir, "test_data.npz")

# Create sample data with distinct dimensions
# rho, u, E have shape (time_steps, spatial_points, num_params)
time_steps = 2
spatial_points = 4
num_params = 3

rho = np.random.randn(time_steps, spatial_points, num_params).astype(np.double)
u = np.random.randn(time_steps, spatial_points, num_params).astype(np.double)
E = np.random.randn(time_steps, spatial_points, num_params).astype(np.double)
x = np.linspace(0, 1, spatial_points).astype(np.double)
params = np.array([0.1, 0.2, 0.3]).astype(np.double)
t = np.linspace(0, 1, time_steps).astype(np.double)

# Save to npz file
np.savez(npz_file, rho, u, E, x, params, t)

ARGS = {
    "data_dir": npz_file,
    "param": 0
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 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
FIB_DAT Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/network.py
code served (permissive licence) · get_code("a337897078dd52eb")
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 2, dtype family float, shape (2, 401) 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 pandas as pd
import numpy as np
import tempfile
import os

# Create a temporary TSV file with the expected structure
temp_dir = tempfile.mkdtemp()
data_file = os.path.join(temp_dir, 'test_data.tsv')

# Create sample data: need shape that when reshaped to (end, 401) works
# If end = data.shape[0] // 401, then data.shape[0] should be divisible by 401
# Let's use 802 rows (end = 2), with 3 columns
num_rows = 802
num_cols = 3

# Create data with columns: x, h, and an index column (column 2)
data_content = []
for i in range(num_rows):
    x_val = float(i)
    h_val = float(i * 2)
    idx_val = i // 401  # index column (column 2)
    data_content.append(f"{x_val}\t{h_val}\t{idx_val}")

with open(data_file, 'w') as f:
    f.write('\n'.join(data_content))

ARGS = {
    'data_dir': data_file,
    'param': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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 (2, 401) — shape (2, 401) 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
POD1 Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/pod.py
code served (permissive licence) · get_code("41d6318bb337174e")
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. 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 numpy as np
import scipy.linalg

# Create a small matrix U with shape (10, 5)
# This will be sliced and used for POD analysis
np.random.seed(0)
U = np.random.randn(10, 5)

# Define slice indices
s_ind = 1  # start index
e_ind = 8  # end index

# Number of modes to extract
modes = 3

ARGS = {
    "U": U,
    "s_ind": s_ind,
    "e_ind": e_ind,
    "modes": modes
}
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 tuple — a tuple output has no array shape of its own; none is 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
POD2 Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/pod.py
code served (permissive licence) · get_code("338d3dc45d0e0fec")
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. 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 numpy as np
import scipy.linalg

# Create a small dataset with shape (time_steps, height, width, parameters)
# Using distinct dimensions: 10 time steps, 4 height, 3 width, 2 parameters
np.random.seed(0)
data = np.random.randn(10, 4, 3, 2).astype(np.float64)

# Define slice indices
s_ind = 0
e_ind = 8

# Number of modes to extract
modes = 3

ARGS = {
    "data": data,
    "s_ind": s_ind,
    "e_ind": e_ind,
    "modes": modes,
}
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 tuple — a tuple output has no array shape of its own; none is 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
POD3 Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/pod.py
code served (permissive licence) · get_code("90ec5a4abf772bdc")
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. 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 numpy as np
import scipy.linalg

# Create small test data
# data shape: (time_steps, spatial_points, variables)
# We need at least 3 variables (indices 0, 1, 2)
np.random.seed(0)

time_steps = 10
spatial_points = 4
num_variables = 3

# Create data array with shape (10, 4, 3)
data = np.random.randn(time_steps, spatial_points, num_variables)

# Define start and end indices for slicing
s_ind = 1
e_ind = 8

# Define number of modes to extract
modes = 2

ARGS = {
    "data": data,
    "s_ind": s_ind,
    "e_ind": e_ind,
    "modes": modes
}
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 tuple — a tuple output has no array shape of its own; none is 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
VKS_DAT Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/network.py
code served (permissive licence) · get_code("477157a7a11b137b")
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 4, dtype family float, shape (3, 4, 5, 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 pickle
import tempfile
import os

# Create a temporary pickle file with sample data
temp_dir = tempfile.mkdtemp()
data_file = os.path.join(temp_dir, "vks_data.pkl")

# Create sample data: shape (4, 5, 3) - will be moved to (3, 4, 5) then sliced to [:, :, :, :2]
# After moveaxis(vks, 2, 0), shape becomes (3, 4, 5)
# After [:, :, :, :2], we need the last dimension to be at least 2
# So original data should have last dim >= 2, let's make it (4, 5, 3, 2) originally
sample_data = np.random.randn(4, 5, 3, 2).astype(np.float32)
# Add some NaN values to test nan_to_num
sample_data[0, 0, 0, 0] = np.nan
sample_data[1, 2, 1, 1] = np.nan

with open(data_file, 'wb') as f:
    pickle.dump(sample_data, f)

ARGS = {
    'data_file': data_file,
    'param': None
}
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 4, dtype family float, shape (3, 4, 5, 2) — shape (3, 4, 5, 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
count_parameters Ran
ran, honoured the contract we drafted
justinbakermath/pod_hbnode/pod_hbnode/param/models.py
code served (permissive licence) · get_code("f6b944f50d3f15ae")
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 int, ndim 0, dtype family int, shape () 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 torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, 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
data_reconstruct Ran
ran on a synthesized input
justinbakermath/pod_hbnode/pod_hbnode/visualization.py
code served (permissive licence) · get_code("3bc4a1ef53bc3e23")
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 Figure. 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 numpy as np
import tempfile
import os

class Args:
    def __init__(self, dataset, model, out_dir, verbose=False):
        self.dataset = dataset
        self.model = model
        self.out_dir = out_dir
        self.verbose = verbose
        self.param = 0

# Create temporary directory for output
temp_dir = tempfile.mkdtemp()

# Test with VKS dataset
# data shape: (time, x, y, channels) where channels=2
vks_data = np.random.randn(3, 5, 6, 2).astype(np.float32)

vks_args = Args(dataset='VKS', model='test_model', out_dir=temp_dir, verbose=False)

ARGS = {
    'data': vks_data,
    'time': 1,
    'args': vks_args
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Figure — a Figure output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Figure 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
normal_kl Ran
ran; our contract draft was wrong, not the code
justinbakermath/pod_hbnode/pod_hbnode/vae/models.py
code served (permissive licence) · get_code("01efe1f42dca1548")
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 (2, 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 torch

torch.manual_seed(0)

# The function normal_kl computes KL divergence between two normal distributions.
# It takes 4 parameters: mu1, lv1, mu2, lv2
# All are used in arithmetic operations (exp, division, subtraction, etc.)
# They should be tensors of compatible shapes for broadcasting

# Create small tensors with batch size 2 and dimension 4
mu1 = torch.randn(2, 4)
lv1 = torch.randn(2, 4)
mu2 = torch.randn(2, 4)
lv2 = torch.randn(2, 4)

ARGS = {
    "mu1": mu1,
    "lv1": lv1,
    "mu2": mu2,
    "lv2": lv2,
}
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 (2, 4) — shape (2, 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
kpp_animate Not yet run
unverified (harvested, no recorded run)
justinbakermath/pod_hbnode/pod_hbnode/visualization.py
code served (permissive licence) · get_code("239655341b0cc933")
How it ran

no invocation recorded.

train Not yet run
unverified (harvested, no recorded run)
justinbakermath/pod_hbnode/pod_hbnode/dmd_model.py
code served (permissive licence) · get_code("95abbedc21d8e2ce")
How it ran

no invocation recorded.

vks_animate Not yet run
unverified (harvested, no recorded run)
justinbakermath/pod_hbnode/pod_hbnode/visualization.py
code served (permissive licence) · get_code("3b2ff7698c4bc6ad")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 2 later) and 13 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 180 other papers
55 dated earlier, 3 the same month, 121 later, 1 same year, order not determinable · 8 only as an assembly brick
1705.08415 (2017-05)
Supervised Community Detection with Line Graph Neural Networks
1707.01926 (2017-07)
Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting
1808.09367 (2018-08)
Deriving Machine Attention from Human Rationales
and 177 more not named here
normal_kl 4 other papers
2 dated earlier, 2 later
2102.02248 (2021-02)
Reduced-order Model for Fluid Flows via Neural Ordinary Differential Equations
2107.11350 (2021-07)
Heteroscedastic Temporal Variational Autoencoder for Irregular Time Series
2312.06015 (2023-12)
Speeding up astrochemical reaction networks with autoencoders and neural ODEs
and 1 more not named here

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

Proper orthogonal decomposition (POD) allows reduced-order modeling of complex dynamical systems at a substantial level, while maintaining a high degree of accuracy in modeling the underlying dynamical systems. Advances in machine learning algorithms enable learning POD-based dynamics from data and making accurate and fast predictions of dynamical systems. This paper extends the recently proposed heavy-ball neural ODEs (HBNODEs) [Xia et al. NeurIPS, 2021] for learning data-driven reduced-order models (ROMs) in the POD context, in particular, for learning dynamics of time-varying coefficients generated by the POD analysis on training snapshots constructed by solving full-order models. HBNODE enjoys several practical advantages for learning POD-based ROMs with theoretical guarantees, including 1) HBNODE can learn long-range dependencies effectively from sequential observations, which is crucial for learning intrinsic patterns from sequential data, and 2) HBNODE is computationally efficient in both training and testing. We compare HBNODE with other popular ROMs on several complex dynamical systems, including the von Kármán Street flow, the Kurganov-Petrova-Popov equation, and the one-dimensional Euler equations for fluids modeling.

For agents

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

get_harvested_code_for_paper("2202.12373")
get_code_for_paper("2202.12373")
have("2202.12373")

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

curl https://syntology.ai/api/ran/2202.12373.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.12373.svg)](https://syntology.ai/paper/2202.12373)

Connect an agent — have() is free.