SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.11283 · 2024

Deep adaptive sampling for surrogate modeling without labeled data

arXiv · PDF · Open in the Atlas

Code that ran

10 of 16 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 8 failures of our instrument; 6 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/2402.11283.json · badge: /api/ran/2402.11283.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
MJfadeaway/DAS-2 canonical (classifier label)
classifier basis: census
10 ran · 6 no recorded run
FunctionStatusWhere it lives
entropy_loss_func Ran
ran; our contract draft was wrong, not the code
MJfadeaway/DAS-2/Lid-driven_cavity_flow/das_fixed.py
code served (permissive licence) · get_code("9cac92f6484b1d10")
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. 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
import numpy as np

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 1)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()

X = torch.randn(2, 4)

Quantity = torch.rand(2, 1)

Pre_Pdf = torch.rand(2, 1) + 0.1

ARGS = {
    "Model": model,
    "X": X,
    "Quantity": Quantity,
    "Pre_Pdf": Pre_Pdf
}
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
entropy_loss_func Ran
ran; our contract draft was wrong, not the code
MJfadeaway/DAS-2/Lid-driven_cavity_flow/das_train.py
code served (permissive licence) · get_code("3b172cb085b5ae82")
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. 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
import numpy as np

torch.manual_seed(0)

# Define a simple Model that takes X and returns log_pdf values
class SimpleModel(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, X):
        return self.linear(X)

# Create model instance
model = SimpleModel(input_dim=4, output_dim=3)

# Create X: batch size 2, feature dim 4
X = torch.randn(2, 4)

# Create Quantity: should match output shape of Model(X), which is (2, 3)
Quantity = torch.randn(2, 3)

# Create Pre_Pdf: should match output shape of Model(X), which is (2, 3)
Pre_Pdf = torch.rand(2, 3) + 0.1  # Add 0.1 to avoid division by zero after scaling

ARGS = {
    "Model": model,
    "X": X,
    "Quantity": Quantity,
    "Pre_Pdf": Pre_Pdf
}
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
entropy_loss_func Ran
ran; our contract draft was wrong, not the code
MJfadeaway/DAS-2/Operator_learning/das_oplearning.py
code served (permissive licence) · get_code("d7b357b308f990f7")
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. 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)

# Define a simple model that takes input X and outputs log_pdf values
class SimpleModel(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, x):
        return self.linear(x)

# Create model instance
model = SimpleModel(input_dim=4, output_dim=3)

# Create input tensor: batch size 2, input dimension 4
X = torch.randn(2, 4)

# Create Quantity tensor: batch size 2, output dimension 3
Quantity = torch.randn(2, 3).abs() + 0.1  # ensure positive values

# Create Pre_Pdf tensor: batch size 2, output dimension 3
Pre_Pdf = torch.randn(2, 3).abs() + 0.1  # ensure positive values to avoid division issues

ARGS = {
    "Model": model,
    "X": X,
    "Quantity": Quantity,
    "Pre_Pdf": Pre_Pdf
}
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
entropy_loss_func Ran
ran; our contract draft was wrong, not the code
MJfadeaway/DAS-2/Optimal_control/das_train.py
code served (permissive licence) · get_code("6d7f5b45c76f2dfa")
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. 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
import numpy as np

torch.manual_seed(0)

# Define a simple model that takes input X and outputs log_pdf values
class SimpleModel(nn.Module):
    def __init__(self, input_dim, output_dim):
        super(SimpleModel, self).__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, x):
        return self.linear(x)

# Create model instance
model = SimpleModel(input_dim=4, output_dim=3)

# Create input tensor X: batch_size=2, feature_dim=4
X = torch.randn(2, 4)

# Create Quantity tensor: batch_size=2, output_dim=3
Quantity = torch.rand(2, 3) + 0.1

# Create Pre_Pdf tensor: batch_size=2, output_dim=3
Pre_Pdf = torch.rand(2, 3) + 0.1

ARGS = {
    "Model": model,
    "X": X,
    "Quantity": Quantity,
    "Pre_Pdf": Pre_Pdf
}
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
flatten_sum Ran
ran; our contract draft was wrong, not the code
MJfadeaway/DAS-2/Lid-driven_cavity_flow/BR_lib/BR_data.py
code served (permissive licence) · get_code("b924ea760370518c")
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, 1) 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 expects a 2D tensor of log probabilities
# It sums along dimension 1 (columns) keeping dimensions
logps = torch.randn(2, 4)

ARGS = {
    "logps": logps
}
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, 1) — shape (2, 1) 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
length_factor Ran
ran, honoured the contract we drafted
MJfadeaway/DAS-2/Optimal_control/das_train.py
code served (permissive licence) · get_code("008dca1dbd361ccd")
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, 1) 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)

# The function expects a tensor with shape (batch_size, 3+)
# It indexes columns 0, 1, and 2
# Batch size: 2, Features: 4 (to have at least 3 columns indexed)
Input = torch.randn(2, 4)

ARGS = {
    "Input": Input
}
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, 1) — shape (2, 1) 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
loss_func Ran
ran; our contract draft was wrong, not the code
MJfadeaway/DAS-2/Lid-driven_cavity_flow/das_train.py
code served (permissive licence) · get_code("c8171b470687f787")
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 torch
import torch.nn as nn

torch.manual_seed(0)

# Create a small network with appropriate architecture
# Input: 3 dimensions (x, y, Re parameter)
# Output: 3 dimensions (u0, u1, p)
net = nn.Module()

class SimpleNet(nn.Module):
    def __init__(self):
        super().__init__()
        self.layers = [3, 8, 8, 3]
        self.activation = nn.Tanh()
        self.linear = nn.ModuleList([nn.Linear(self.layers[i], self.layers[i + 1]) for i in range(len(self.layers) - 1)])
        for i in range(len(self.layers) - 1):
            nn.init.xavier_normal_(self.linear[i].weight.data, gain=1.0)
            nn.init.zeros_(self.linear[i].bias.data)

    def forward(self, x):
        if not torch.is_tensor(x):
            x = torch.from_numpy(x)
        a = self.activation(self.linear[0](x))
        for i in range(1, len(self.layers) - 2):
            z = self.linear[i](a)
            a = self.activation(z)
        a = self.linear[-1](a)
        return a

net = SimpleNet()

# Create interior training set with batch size 2, 3 input dimensions
Inner_x = torch.randn(2, 3, dtype=torch.float32)

# Create boundary set object with required attributes
class BoundarySet:
    def __init__(self):
        self.x = torch.randn(2, 3, dtype=torch.float32)
        self.r0 = torch.randn(2, 1, dtype=torch.float32)
        self.r1 = torch.randn(2, 1, dtype=torch.float32)
        self.weighted0 = torch.ones(2, 1, dtype=torch.float32)
        self.weighted1 = torch.ones(2, 1, dtype=torch.float32)
        self.u = None
        self.u0 = None
        self.u1 = None
        self.res0 = None
        self.res1 = None

BdSet = BoundarySet()

ARGS = {
    'Net': net,
    'Inner_x': Inner_x,
    'BdSet': BdSet,
    'beta': 1.0
}
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 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
loss_func Ran
ran; our contract draft was wrong, not the code
MJfadeaway/DAS-2/Operator_learning/das_oplearning.py
code served (permissive licence) · get_code("feabd500549c048c")
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 torch
import torch.nn as nn

torch.manual_seed(0)

# Define a minimal Model that satisfies the function's usage
class SimpleModel(nn.Module):
    def __init__(self, input_dim, sensor_dim):
        super().__init__()
        # Model takes CollocationPoints (1D) and SensorValues (m-dim)
        # and outputs a scalar prediction
        self.linear1 = nn.Linear(1 + sensor_dim, 16)
        self.linear2 = nn.Linear(16, 1)
    
    def forward(self, points, sensor_values):
        # points: shape (batch, 1)
        # sensor_values: shape (batch, m)
        # concatenate and pass through network
        combined = torch.cat([points, sensor_values], dim=1)
        x = torch.relu(self.linear1(combined))
        output = self.linear2(x)
        return output

# Create model instance
num_samples = 2  # batch size
m = 4  # sensor dimension
Q = 3  # number of collocation points
sensor_dim = 4

model = SimpleModel(input_dim=1, sensor_dim=sensor_dim)

# Create SensorValues: shape (num, m)
SensorValues = torch.randn(num_samples, m)

# Create CollocationPoints: shape (num, Q)
CollocationPoints = torch.randn(num_samples, Q)

# Create SourceTerm: shape (num, Q)
SourceTerm = torch.randn(num_samples, Q)

# Beta parameter
beta = 100.0

ARGS = {
    'Model': model,
    'SensorValues': SensorValues,
    'CollocationPoints': CollocationPoints,
    'SourceTerm': SourceTerm,
    'beta': beta
}
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 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
uu Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
MJfadeaway/DAS-2/Lid-driven_cavity_flow/data.py
code served (permissive licence) · get_code("87af8f0ef57cfc9e")
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, 1) 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)

# x: coordinates of data points - shape (batch_size, spatial_dim)
# The function accesses x[:, 0:1] and x[:, 1], so we need at least 2 spatial dimensions
x = torch.randn(2, 4)

# ind: index of variable - integer (0 or 1)
ind = 0

# Re: Reynolds number - scalar float
Re = 100.0

ARGS = {
    "x": x,
    "ind": ind,
    "Re": Re
}
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, 1) — shape (2, 1) 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
yd Ran
ran, honoured the contract we drafted
MJfadeaway/DAS-2/Optimal_control/das_train.py
code served (permissive licence) · get_code("640c69f3737dbb7a")
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, 1) 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 numpy as np

torch.manual_seed(0)

# Create a small tensor with shape (batch_size=2, features=4)
# The function indexes columns 0:1, 3:4, so we need at least 4 columns
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 2, dtype family float, shape (2, 1) — shape (2, 1) 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
absolute_error Not yet run
unverified (harvested, no recorded run)
MJfadeaway/DAS-2/Lid-driven_cavity_flow/das_fixed.py
code served (permissive licence) · get_code("d481fa3ca02b9f3a")
How it ran

no invocation recorded.

line_search_wolfe1 Not yet run
unverified (harvested, no recorded run)
MJfadeaway/DAS-2/Lid-driven_cavity_flow/linesearch.py
code served (permissive licence) · get_code("fd0ebec955ca8a6d")
How it ran

no invocation recorded.

log_standard_Gaussian Not yet run
unverified (harvested, no recorded run)
MJfadeaway/DAS-2/Lid-driven_cavity_flow/BR_lib/BR_data.py
code served (permissive licence) · get_code("9762ab5f23f45a06")
How it ran

no invocation recorded.

loss_func Not yet run
unverified (harvested, no recorded run)
MJfadeaway/DAS-2/Lid-driven_cavity_flow/das_fixed.py
code served (permissive licence) · get_code("2767233d06d2a32a")
How it ran

no invocation recorded.

relative_error Not yet run
unverified (harvested, no recorded run)
MJfadeaway/DAS-2/Lid-driven_cavity_flow/das_train.py
code served (permissive licence) · get_code("5c13cffdbbc3ac03")
How it ran

no invocation recorded.

scalar_search_wolfe1 Not yet run
unverified (harvested, no recorded run)
MJfadeaway/DAS-2/Lid-driven_cavity_flow/linesearch.py
code served (permissive licence) · get_code("64a3d2b824b7001e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 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 16 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

Surrogate modeling is of great practical significance for parametric differential equation systems. In contrast to classical numerical methods, using physics-informed deep learning methods to construct simulators for such systems is a promising direction due to its potential to handle high dimensionality, which requires minimizing a loss over a training set of random samples. However, the random samples introduce statistical errors, which may become the dominant errors for the approximation of low-regularity and high-dimensional problems. In this work, we present a deep adaptive sampling method for surrogate modeling ($\text{DAS}^2$), where we generalize the deep adaptive sampling (DAS) method [62] [Tang, Wan and Yang, 2023] to build surrogate models for low-regularity parametric differential equations. In the parametric setting, the residual loss function can be regarded as an unnormalized probability density function (PDF) of the spatial and parametric variables. This PDF is approximated by a deep generative model, from which new samples are generated and added to the training set. Since the new samples match the residual-induced distribution, the refined training set can further reduce the statistical error in the current approximate solution. We demonstrate the effectiveness of $\text{DAS}^2$ with a series of numerical experiments, including the parametric lid-driven 2D cavity flow problem with a continuous range of Reynolds numbers from 100 to 1000.

For agents

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

get_harvested_code_for_paper("2402.11283")
get_code_for_paper("2402.11283")
have("2402.11283")

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

curl https://syntology.ai/api/ran/2402.11283.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.11283.svg)](https://syntology.ai/paper/2402.11283)

Connect an agent — have() is free.