SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.18563 · ICLR · 2025

No Equations Needed: Learning System Dynamics Without Relying on Closed-Form ODEs

Mihaela Van Der Schaar, Krzysztof Kacprzyk

arXiv · PDF · Open in the Atlas

Code that ran

7 of 7 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 3 violated it and 2 have no contract verdict; 2 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/2501.18563.json · badge: /api/ran/2501.18563.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
krzysztof-kacprzyk/semanticode canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
vanderschaarlab/SemanticODE community
Papers with Code archive
6 ran · 0 no recorded run
FunctionStatusWhere it lives
Knot Ran
ran on a synthesized input · constructed an object; no result was computed
vanderschaarlab/SemanticODE/semantic_odes/trajectory_predictors.py
code served (permissive licence) · get_code("22670b4943d4a85f")
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, Knot: 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 numpy as np

torch.manual_seed(0)

# Create simple tensor arguments for x and y
x = torch.randn(2, 4)
y = torch.randn(2, 3)

# Create a transition_point as a scalar or tensor
transition_point = 0.5

ARGS = {
    "x": x,
    "y": y,
    "transition_point": transition_point,
}
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 Knot — 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 Knot 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
WrappedCubicSpline Ran
ran on a synthesized input · constructed an object; no result was computed
vanderschaarlab/SemanticODE/semantic_odes/trajectory_predictors.py
code served (permissive licence) · get_code("7586a49dea800127")
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, WrappedCubicSpline: 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 numpy as np

class SemanticRepresentation:
    """Stand-in class for semantic_representation parameter"""
    def __init__(self):
        self.composition = ['++c', '+-c', '--c']
        self.coordinates_finite_composition = np.array([
            [0.0, 1.0],
            [1.0, 2.0],
            [2.0, 1.5],
            [3.0, 0.5]
        ])
        self.t_range = (0.0, 3.0)
        self.derivative_start = 0.5
        self.derivative_end = -0.5
        self.second_derivative_end = 0.1

class MockUtils:
    """Mock utils module with required functions"""
    @staticmethod
    def is_unbounded_composition(composition):
        if len(composition) == 0:
            return False
        motif = composition[-1]
        return motif[2] != 'c'

class MockWrappedCubicSpline:
    """Mock WrappedCubicSpline with required methods and attributes"""
    def type_of_transition_point(self, i):
        transition_types = ['start', 'inflection', 'max', 'end']
        return transition_types[min(i, len(transition_types) - 1)]
    
    @property
    def first_derivative_sign(self):
        return [1, -1, 1]

# Create a mock utils module
utils = MockUtils()

# Create semantic representation
semantic_rep = SemanticRepresentation()

ARGS = {
    'semantic_representation': semantic_rep,
    'epsilon': 1e-09,
    'epsilon_ratio': 1e-06
}
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 WrappedCubicSpline — 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 WrappedCubicSpline 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_full_composition_library Ran
ran; our contract draft was wrong, not the code
krzysztof-kacprzyk/semanticode/semantic_odes/api.py
code served (permissive licence) · get_code("f47091a804873629")
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 list. 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.

# Setup code for create_full_composition_library

ARGS = {
    'max_length': 3,
    'is_infinite': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list 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
is_bounded_motif Ran
ran, violated the contract we drafted
vanderschaarlab/SemanticODE/semantic_odes/trajectory_predictors.py
code served (permissive licence) · get_code("c13f060e770fa143")
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 bool, 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 numpy as np

# The function is_bounded_motif takes a motif string and checks if the character
# at index 2 is 'c'. We need to provide a valid string argument.

# Create a motif string where index 2 is 'c' (bounded case)
motif_bounded = "abcdef"

# Alternatively, we could test with a motif where index 2 is not 'c' (unbounded case)
# For now, let's use the bounded case
motif = "abcdef"

ARGS = {
    "motif": motif
}
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 bool, 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
is_unbounded_composition Ran
ran, violated the contract we drafted
vanderschaarlab/SemanticODE/semantic_odes/trajectory_predictors.py
code served (permissive licence) · get_code("5219e152da820761")
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 bool, 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 numpy as np

# The function is_unbounded_composition takes a composition parameter
# which is a list (or tuple) of strings representing motifs.
# 
# Looking at the function body:
# - It calls len(composition) to get n_motifs
# - It accesses composition[n_motifs - 1] to get the last motif
# - It passes that motif to is_unbounded_motif
# 
# is_unbounded_motif calls is_bounded_motif(motif)
# is_bounded_motif checks if motif[2] == 'c'
# 
# So each string in composition must be at least 3 characters long,
# and the character at index 2 determines bounded/unbounded status.

# Create a valid composition: a list of motif strings
# Each motif needs to be at least 3 characters long
# We'll create motifs where the 3rd character (index 2) determines boundedness

composition = [
    "abc",      # bounded (motif[2] == 'c')
    "abx",      # unbounded (motif[2] != 'c')
    "defg"      # bounded (motif[2] == 'f', but we need 'c' at index 2)
]

# Actually, let me reconsider - motif[2] means index 2, so third character
# Let's make it clearer:
composition = [
    "xxc",      # bounded (index 2 is 'c')
    "aac",      # bounded (index 2 is 'c')
    "xxy"       # unbounded (index 2 is 'y', not 'c')
]

ARGS = {
    "composition": composition
}
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 bool, 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
is_unbounded_motif Ran
ran, violated the contract we drafted
vanderschaarlab/SemanticODE/semantic_odes/trajectory_predictors.py
code served (permissive licence) · get_code("9cb5cec659e4ed5b")
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 bool, 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 numpy as np

torch.manual_seed(0)

# The function is_unbounded_motif takes a motif parameter
# Looking at the function body, it calls is_bounded_motif(motif)
# which accesses motif[2] and compares it to 'c'
# So motif must be a string with at least 3 characters

# Create a test motif string
# According to the logic:
# - if motif[2] == 'c', it's bounded (is_bounded_motif returns True)
# - if motif[2] != 'c', it's unbounded (is_bounded_motif returns False, so is_unbounded_motif returns True)

# Test case 1: bounded motif (has 'c' at index 2)
bounded_motif = "abc"

# Test case 2: unbounded motif (does not have 'c' at index 2)
unbounded_motif = "abd"

ARGS = {
    "motif": unbounded_motif
}
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 bool, 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
type_of_transition_point Ran
ran; our contract draft was wrong, not the code
vanderschaarlab/SemanticODE/semantic_odes/trajectory_predictors.py
code served (permissive licence) · get_code("c2a4934fc75b7191")
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 str. 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.

# Setup code for type_of_transition_point function

# Test case 1: start transition point
composition_1 = ['++up', '+-down', '--flat']
ind_1 = 0

# Test case 2: end transition point
composition_2 = ['++up', '+-down', '--flat']
ind_2 = 3

# Test case 3: inflection point (++ to +-)
composition_3 = ['++up', '+-mixed', '--flat']
ind_3 = 1

# Test case 4: max point (+- to --)
composition_4 = ['++up', '+-mixed', '--down']
ind_4 = 2

# Test case 5: min point (-+ to ++)
composition_5 = ['-+bottom', '++up', '--down']
ind_5 = 1

# Define ARGS with a valid test case
ARGS = {
    'composition': ['++up', '+-mixed', '--down'],
    'ind': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 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 7 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

Data-driven modeling of dynamical systems is a crucial area of machine learning. In many scenarios, a thorough understanding of the model's behavior becomes essential for practical applications. For instance, understanding the behavior of a pharmacokinetic model, constructed as part of drug development, may allow us to both verify its biological plausibility (e.g., the drug concentration curve is nonnegative and decays to zero in the long term) and to design dosing guidelines (e.g., by looking at the peak concentration and its timing). Discovery of closed-form ordinary differential equations (ODEs) can be employed to obtain such insights by finding a compact mathematical equation and then analyzing it (a two-step approach). However, its widespread use is currently hindered because the analysis process may be time-consuming, requiring substantial mathematical expertise, or even impossible if the equation is too complex. Moreover, if the found equation's behavior does not satisfy the requirements, editing it or influencing the discovery algorithms to rectify it is challenging as the link between the symbolic form of an ODE and its behavior can be elusive. This paper proposes a conceptual shift to modeling low-dimensional dynamical systems by departing from the traditional two-step modeling process. Instead of first discovering a closed-form equation and then analyzing it, our approach, direct semantic modeling, predicts the semantic representation of the dynamical system (i.e., description of its behavior) directly from data, bypassing the need for complex post-hoc analysis. This direct approach also allows the incorporation of intuitive inductive biases into the optimization algorithm and editing the model's behavior directly, ensuring that the model meets the desired specifications. Our approach not only simplifies the modeling pipeline but also enhances the transparency and flexibility of the resulting models compared to traditional closed-form ODEs.

For agents

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

get_harvested_code_for_paper("2501.18563")
get_code_for_paper("2501.18563")
have("2501.18563")

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

curl https://syntology.ai/api/ran/2501.18563.json

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

[![Syntology run record](https://syntology.ai/api/ran/2501.18563.svg)](https://syntology.ai/paper/2501.18563)

Connect an agent — have() is free.