SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.04084 · 2022

Adaptive Self-supervision Algorithms for Physics-informed Neural Networks

arXiv · PDF · Open in the Atlas

Code that ran

7 of 9 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 7 have no contract verdict; 0 failures 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/2207.04084.json · badge: /api/ran/2207.04084.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
shashanksubramanian/adaptive-selfsupervision-pinns canonical (classifier label)
classifier basis not recorded on the link
7 ran · 2 no recorded run
FunctionStatusWhere it lives
create_ic Ran
ran on a synthesized input
shashanksubramanian/adaptive-selfsupervision-pinns/utils/pde_sols.py
code served (permissive licence) · get_code("cbc90fa0d1ca51bd")
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 function. 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

ARGS = {
    'u_0': 'sin(x)'
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
create_new_training_data Ran
ran on a synthesized input
shashanksubramanian/adaptive-selfsupervision-pinns/utils/data_utils.py
code served (permissive licence) · get_code("bf1049301bf31879")
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

class Params:
    def __init__(self, system, enable_obs=False):
        self.system = system
        self.enable_obs = enable_obs

class Domain:
    def __init__(self, system, enable_obs=False):
        self.ic = np.random.rand(2, 4)
        self.bc = np.random.rand(3, 4)
        self.col = np.random.rand(5, 4)
        
        self.bc_left = np.random.rand(2, 4)
        self.bc_right = np.random.rand(2, 4)
        self.bc_bot = np.random.rand(2, 4)
        self.bc_top = np.random.rand(2, 4)
        self.obs = np.random.rand(1, 4)
        
        self.sol = [
            np.random.rand(1, 4),
            np.random.rand(2, 4),
            np.random.rand(3, 4),
            np.random.rand(5, 4),
            np.random.rand(2, 4),
            np.random.rand(2, 4),
            np.random.rand(1, 4)
        ]

params_non_poisson = Params(system='burgers', enable_obs=False)
domain_non_poisson = Domain(system='burgers', enable_obs=False)

params_poisson = Params(system='poisson', enable_obs=False)
domain_poisson = Domain(system='poisson', enable_obs=False)

params_poisson_obs = Params(system='poisadv', enable_obs=True)
domain_poisson_obs = Domain(system='poisadv', enable_obs=True)

ARGS = {
    'params': params_poisson,
    'domain': domain_poisson
}
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
fake_mask Ran
ran on a synthesized input
shashanksubramanian/adaptive-selfsupervision-pinns/utils/domains.py
code served (permissive licence) · get_code("b843cfc6fbc377f1")
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 (64, 48) 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

# The function fake_mask takes three parameters:
# - nt: number of time steps (int)
# - nx: number of spatial points (int)
# - n: normalization factor (int, default 100)
#
# The function creates a 2D mask array of shape (nt, nx),
# sets a rectangular region to 1, counts nonzero elements,
# and normalizes by dividing by that count.

ARGS = {
    "nt": 64,
    "nx": 48,
    "n": 100,
}
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 2, dtype family float, shape (64, 48) — shape (64, 48) 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
fft_coef Ran
ran on a synthesized input
shashanksubramanian/adaptive-selfsupervision-pinns/utils/misc_utils.py
code served (permissive licence) · get_code("48df2071a578681a")
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 (8,) 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

def fft_coef(n):
    ikx_pos = 1j * np.arange(0, n // 2 + 1, 1)
    ikx_neg = 1j * np.arange(-n // 2 + 1, 0, 1)
    ikx = np.concatenate((ikx_pos, ikx_neg))
    return ikx

ARGS = {
    "n": 8
}
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 (8,) — shape (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
get_diff_tensor Ran
ran on a synthesized input
shashanksubramanian/adaptive-selfsupervision-pinns/utils/misc_utils.py
code served (permissive licence) · get_code("9b5d5ccfb3368d0e")
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 dict. 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 numpy as np

torch.manual_seed(0)

ARGS = {
    'diff_type': 'iso'
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_velocity Ran
ran on a synthesized input
shashanksubramanian/adaptive-selfsupervision-pinns/utils/misc_utils.py
code served (permissive licence) · get_code("0c9b4a2f9ecd0f92")
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 dict. 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

torch.manual_seed(0)

ARGS = {
    'vel_type': 'iso'
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
set_scheduler Ran
ran on a synthesized input
shashanksubramanian/adaptive-selfsupervision-pinns/utils/optimizer_utils.py
code served (permissive licence) · get_code("8fcad789ad4c0740")
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 ReduceLROnPlateau. 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
from torch.optim import Adam

torch.manual_seed(0)

# Create a minimal model
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Create model and optimizer
model = MinimalModel()
optimizer = Adam(model.parameters(), lr=0.001)

# Create args object with required attributes
class ArgsConfig:
    def __init__(self, scheduler_type, patience=5, max_epochs=100):
        self.scheduler = scheduler_type
        self.patience = patience
        self.max_epochs = max_epochs

ARGS = {
    'args': ArgsConfig(scheduler_type='reducelr', patience=5, max_epochs=100),
    'optimizer': optimizer
}
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 ReduceLROnPlateau — a ReduceLROnPlateau output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ReduceLROnPlateau output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_data_loader Not yet run
unverified (harvested, no recorded run)
shashanksubramanian/adaptive-selfsupervision-pinns/utils/data_utils.py
code served (permissive licence) · get_code("f0c9ddfca5da08e9")
How it ran

no invocation recorded.

resample_data_loader Not yet run
unverified (harvested, no recorded run)
shashanksubramanian/adaptive-selfsupervision-pinns/utils/data_utils.py
code served (permissive licence) · get_code("4fcbeb29563cd648")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 9 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 2026-11-21.

Repositories linked to this paper

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

Abstract

Physics-informed neural networks (PINNs) incorporate physical knowledge from the problem domain as a soft constraint on the loss function, but recent work has shown that this can lead to optimization difficulties. Here, we study the impact of the location of the collocation points on the trainability of these models. We find that the vanilla PINN performance can be significantly boosted by adapting the location of the collocation points as training proceeds. Specifically, we propose a novel adaptive collocation scheme which progressively allocates more collocation points (without increasing their number) to areas where the model is making higher errors (based on the gradient of the loss function in the domain). This, coupled with a judicious restarting of the training during any optimization stalls (by simply resampling the collocation points in order to adjust the loss landscape) leads to better estimates for the prediction error. We present results for several problems, including a 2D Poisson and diffusion-advection system with different forcing functions. We find that training vanilla PINNs for these problems can result in up to 70% prediction error in the solution, especially in the regime of low collocation points. In contrast, our adaptive schemes can achieve up to an order of magnitude smaller error, with similar computational complexity as the baseline. Furthermore, we find that the adaptive methods consistently perform on-par or slightly better than vanilla PINN method, even for large collocation point regimes. The code for all the experiments has been open sourced.

For agents

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

get_harvested_code_for_paper("2207.04084")
get_code_for_paper("2207.04084")
have("2207.04084")

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

curl https://syntology.ai/api/ran/2207.04084.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.04084.svg)](https://syntology.ai/paper/2207.04084)

Connect an agent — have() is free.