SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.10642 · ICML · 2024

Using Uncertainty Quantification to Characterize and Improve Out-of-Domain Learning for PDEs

Michael Mahoney, Danielle Maddix, Yuyang Wang, S Chandra Mouli, Gaurav Gupta, Shima Alizadeh, Andrew Stuart

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2403.10642.json · badge: /api/ran/2403.10642.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
amazon-science/operator-probconserv canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
apply_constraint Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
amazon-science/operator-probconserv/probconserv.py
code served (permissive licence) · get_code("b38988d1fd23111f")
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 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 numpy as np
from torch.distributions import Normal

torch.manual_seed(0)

# Define dimensions with distinct sizes
nf = 2  # number of functions/samples
nx = 4  # number of spatial points
nt = 3  # number of time points
nt_full = 5  # full time dimension

# Create time array
t = torch.linspace(0, 1, nt_full)

# Create spatial grid
grid_train = torch.linspace(0, 1, nx)

# Create mu: shape (nf, nx, nt)
mu = torch.randn(nf, nx, nt)

# Create std: shape (nf, nx, nt)
std = torch.abs(torch.randn(nf, nx, nt)) + 0.1

# Define mass_rhs_func that takes inputs of shape (nf, nt, nx, 2) and returns (nf, nt)
def mass_rhs_func(inputs):
    """
    inputs: shape (nf, nt, nx, 2) where last dim is (t, x)
    returns: shape (nf, nt)
    """
    nf_in, nt_in, nx_in, _ = inputs.shape
    # Simple function: sum over spatial dimension
    return torch.sum(inputs[:, :, :, 1], dim=2)  # shape (nf, nt)

# Define time prediction slice
tpred = (0, nt)  # slice indices for time dimension

# Precision parameter
precis_g = np.inf

# Second derivative alpha (set to None to skip second derivative constraint)
second_deriv_alpha = None

ARGS = {
    'mu': mu,
    'std': std,
    'mass_rhs_func': mass_rhs_func,
    't': t,
    'tpred': tpred,
    'grid_train': grid_train,
    'precis_g': precis_g,
    'second_deriv_alpha': second_deriv_alpha,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
einops, numpy, 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
get_empirical_mass_rhs Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
amazon-science/operator-probconserv/probconserv.py
code served (permissive licence) · get_code("2eef5415e373e5fd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from einops import rearrange, reduce, repeat

torch.manual_seed(0)

# The function expects 'outs' which is indexed as outs[:, 1:] and outs[:, :-1]
# This suggests a tensor with shape (nf, nx, nt) where:
# - nf: number of features
# - nx: spatial dimension (needs at least 2 elements to slice [1:] and [:-1])
# - nt: time dimension

# Create a small tensor with distinct dimensions
# nf=2, nx=4, nt=3
outs = torch.randn(2, 4, 3)

ARGS = {
    "outs": outs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
einops, 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, 3) — shape (2, 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
meshgrid Ran
ran; our contract draft was wrong, not the code
amazon-science/operator-probconserv/probconserv.py
code served (permissive licence) · get_code("e5586d2c62ef0a99")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 4, dtype family float, shape (2, 3, 4, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from einops import rearrange, repeat

torch.manual_seed(0)

# Create small tensors with distinct dimensions
# ts: shape (nf, nt) where nf=2 (num features), nt=3 (num time steps)
ts = torch.randn(2, 3)

# xs: shape (nf, nx) where nf=2 (num features), nx=4 (num spatial points)
xs = torch.randn(2, 4)

ARGS = {
    "ts": ts,
    "xs": xs,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
einops, 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 4, dtype family float, shape (2, 3, 4, 2) — shape (2, 3, 4, 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

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

Abstract

Existing work in scientific machine learning (SciML) has shown that data-driven learning of solution operators can provide a fast approximate alternative to classical numerical partial differential equation (PDE) solvers. Of these, Neural Operators (NOs) have emerged as particularly promising. We observe that several uncertainty quantification (UQ) methods for NOs fail for test inputs that are even moderately out-of-domain (OOD), even when the model approximates the solution well for in-domain tasks. To address this limitation, we show that ensembling several NOs can identify high-error regions and provide good uncertainty estimates that are well-correlated with prediction errors. Based on this, we propose a cost-effective alternative, DiverseNO, that mimics the properties of the ensemble by encouraging diverse predictions from its multiple heads in the last feed-forward layer. We then introduce Operator-ProbConserv, a method that uses these well-calibrated UQ estimates within the ProbConserv framework to update the model. Our empirical results show that Operator-ProbConserv enhances OOD model performance for a variety of challenging PDE problems and satisfies physical constraints such as conservation laws.

For agents

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

get_harvested_code_for_paper("2403.10642")
get_code_for_paper("2403.10642")
have("2403.10642")

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

curl https://syntology.ai/api/ran/2403.10642.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.10642.svg)](https://syntology.ai/paper/2403.10642)

Connect an agent — have() is free.