SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.06366 · NeurIPS · 2024

Physics-Informed Regularization for Domain-Agnostic Dynamical System Modeling

Yizhou Sun, Wei Wang, Zijie Huang, Ziniu Hu, Xiao Luo, Yuanzhou Chen, Wanjia Zhao, Jingdong Gao, Yadi Cao, Los Angeles

arXiv · PDF · Open in the Atlas

Code that ran

4 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 7 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/2410.06366.json · badge: /api/ran/2410.06366.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
wanjiaZhao1203/TREAT canonical (classifier label)
classifier basis: classifier_2025
4 ran · 7 no recorded run
FunctionStatusWhere it lives
GraphODEFuncT Ran
ran on a synthesized input · constructed an object; no result was computed
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("9bfbecf4c1cf363c")
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 an instance of its own class, GraphODEFuncT: the run constructed the object and did not call it, so no result was computed. 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)

# Build a minimal ode_func_net that satisfies the usage in GraphODEFuncT
class MinimalGraphConvLayer(nn.Module):
    def __init__(self):
        super(MinimalGraphConvLayer, self).__init__()
        self.base_conv = nn.Module()
        self.base_conv.rel_type = None
        self.base_conv.rel_rec = None
        self.base_conv.rel_send = None
        self.base_conv.edge_types = None

class MinimalODEFuncNet(nn.Module):
    def __init__(self, input_dim=4, output_dim=4):
        super(MinimalODEFuncNet, self).__init__()
        self.gcs = nn.ModuleList([
            MinimalGraphConvLayer(),
            MinimalGraphConvLayer()
        ])
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, z):
        # z should be a tensor of shape (batch, dim)
        return self.linear(z)

# Create the ode_func_net
ode_func_net = MinimalODEFuncNet(input_dim=4, output_dim=4)

# Define ARGS
ARGS = {
    'ode_func_net': ode_func_net
}
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 GraphODEFuncT — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GraphODEFuncT 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
HamiltionEquation Ran
ran on a synthesized input · constructed an object; no result was computed
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("29f52a3c2b924713")
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 an instance of its own class, HamiltionEquation: the run constructed the object and did not call it, so no result was computed. 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)

# Parameters for HamiltionEquation constructor
ARGS = {
    'dim': 2,
    'time_augment': False,
    'nb_units': 16,
    'nb_layers': 2,
    'activation': 'tanh'
}
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 HamiltionEquation — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a HamiltionEquation 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
ODEFunc Ran
ran on a synthesized input · constructed an object; no result was computed
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("db629ae66a53ba66")
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 an instance of its own class, ODEFunc: the run constructed the object and did not call it, so no result was computed. 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 = {
    'dim': 4,
    'time_augment': True,
    'nb_units': 64,
    'nb_layers': 2,
    'activation': 'tanh'
}
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 ODEFunc — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ODEFunc 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
TemporalEncoding Ran
ran on a synthesized input · constructed an object; no result was computed
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("7f8c3af0e536def3")
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 an instance of its own class, TemporalEncoding: the run constructed the object and did not call it, so no result was computed. 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
import numpy as np

torch.manual_seed(0)

ARGS = {
    "d_hid": 4
}
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 TemporalEncoding — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a TemporalEncoding 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
GNN Not yet run
unverified (harvested, no recorded run)
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("34f19b6fac5bf7b7")
How it ran

no invocation recorded.

GTrans Not yet run
unverified (harvested, no recorded run)
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("292ab2a409724d21")
How it ran

no invocation recorded.

GeneralConv Not yet run
unverified (harvested, no recorded run)
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("bf0142ddc3b334db")
How it ran

no invocation recorded.

NRIConv Not yet run
unverified (harvested, no recorded run)
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("afcb3ef3cb2661e1")
How it ran

no invocation recorded.

ODENetwork Not yet run
unverified (harvested, no recorded run)
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("d20f73464483ea6c")
How it ran

no invocation recorded.

append_activation Not yet run
unverified (harvested, no recorded run)
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("51c8d4a7ddd73e22")
How it ran

no invocation recorded.

init_network_weights Not yet run
unverified (harvested, no recorded run)
wanjiaZhao1203/TREAT/lib/trsoden.py
code served (permissive licence) · get_code("1e19df57d2ee1e3a")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 10 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
init_network_weights
brick of this paper's assembly: helper
1 other paper
1 dated earlier · 1 only as an assembly brick
2107.08273 (2021-07)
STRODE: Stochastic Boundary Ordinary Differential Equation

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

Learning complex physical dynamics purely from data is challenging due to the intrinsic properties of systems to be satisfied. Incorporating physics-informed priors, such as in Hamiltonian Neural Networks (HNNs), achieves high-precision modeling for energy-conservative systems. However, real-world systems often deviate from strict energy conservation and follow different physical priors. To address this, we present a framework that achieves high-precision modeling for a wide range of dynamical systems from the numerical aspect, by enforcing Time-Reversal Symmetry (TRS) via a novel regularization term. It helps preserve energies for conservative systems while serving as a strong inductive bias for non-conservative, reversible systems. While TRS is a domain-specific physical prior, we present the first theoretical proof that TRS loss can universally improve modeling accuracy by minimizing higher-order Taylor terms in ODE integration, which is numerically beneficial to various systems regardless of their properties, even for irreversible systems. By integrating the TRS loss within neural ordinary differential equation models, the proposed model TREAT demonstrates superior performance on diverse physical systems. It achieves a significant 11.5% MSE improvement in a challenging chaotic triple-pendulum scenario, underscoring TREAT's broad applicability and effectiveness. Code and further details are available at here.

For agents

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

get_harvested_code_for_paper("2410.06366")
get_code_for_paper("2410.06366")
have("2410.06366")

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

curl https://syntology.ai/api/ran/2410.06366.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.06366.svg)](https://syntology.ai/paper/2410.06366)

Connect an agent — have() is free.