SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.12604 · 2023

Neural Laplace Control for Continuous-time Delayed Systems

arXiv · PDF · Open in the Atlas

Code that ran

13 of 17 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, 1 violated it and 8 have no contract verdict; 4 failures of our instrument; 4 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/2302.12604.json · badge: /api/ran/2302.12604.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
vanderschaarlab/neurallaplacecontrol canonical (classifier label)
classifier basis not recorded on the link
9 ran · 4 no recorded run
samholt/neurallaplacecontrol canonical (classifier label)
classifier basis: classifier_2025
4 ran · 0 no recorded run
FunctionStatusWhere it lives
acrobot_dynamics_dt_delay Ran
ran on a synthesized input
vanderschaarlab/neurallaplacecontrol/oracle.py
code served (permissive licence) · get_code("f6d5dc82ebceee06")
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, 4) 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 state tensor with shape (batch_size=2, state_dim=4)
# state can be either 4D or 6D based on the function logic
state = torch.randn(2, 4)

# Create perturbed_action with shape (batch_size=2, time_steps, action_dim=2)
# The function indexes perturbed_action[:, -(delay + 1), :nu]
# So we need at least (delay + 1) time steps
delay = 2
perturbed_action = torch.randn(2, delay + 3, 2)

# Create ts (timestep) tensor with shape (batch_size=2,) or (batch_size=2, 1)
ts = torch.tensor([[0.01], [0.01]])

ARGS = {
    'state': state,
    'perturbed_action': perturbed_action,
    'ts': ts,
    'delay': delay,
    'ACTION_LOW': -5.0,
    'ACTION_HIGH': 5.0,
    'nu': 2,
    'friction': False,
}
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, 4) — shape (2, 4) 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
cartpole_dynamics_dt_delay Ran
ran on a synthesized input
vanderschaarlab/neurallaplacecontrol/oracle.py
code served (permissive licence) · get_code("bd9d131794b04d99")
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, 4) 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)

# Construct arguments for cartpole_dynamics_dt_delay
batch_size = 2
delay = 3
nu = 1

# state: batch_size x state_dim (4 or 5 dimensions)
# Using state_dim=4 for the simpler case
state = torch.randn(batch_size, 4)

# perturbed_action: batch_size x (delay + 1) x nu
# The function extracts perturbed_action[:, -(delay + 1), :nu]
# So we need at least (delay + 1) time steps in dimension 1
perturbed_action = torch.randn(batch_size, delay + 1, nu)

# ts: time step(s), can be 1D or 2D
ts = torch.tensor([[0.02], [0.02]], dtype=torch.float32)

# delay parameter
delay_param = delay

# Other optional parameters with defaults
ACTION_LOW = -3.0
ACTION_HIGH = 3.0
nu_param = nu
friction = False
friction_cart = 0.0005
friction_pole = 2e-06

ARGS = {
    "state": state,
    "perturbed_action": perturbed_action,
    "ts": ts,
    "delay": delay_param,
    "ACTION_LOW": ACTION_LOW,
    "ACTION_HIGH": ACTION_HIGH,
    "nu": nu_param,
    "friction": friction,
    "friction_cart": friction_cart,
    "friction_pole": friction_pole,
}
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, 4) — shape (2, 4) 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
create_classifier Ran
ran; our contract draft was wrong, not the code
vanderschaarlab/neurallaplacecontrol/baseline_models/latent_ode_lib/base_models.py
pointer only (licence: MIT); the stored copy get_code serves, from BenJMaurel/PharmaNODE, is pointer only (licence: NONE) · get_code("fc9a64735be0911c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from BenJMaurel/PharmaNODE, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

generate_log_file_path Ran
ran; our contract draft was wrong, not the code
samholt/neurallaplacecontrol/run_exp_multi.py
code served (permissive licence) · get_code("a00c4f55aee7a564")
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 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.

import os
import time
from pathlib import Path

# The function uses:
# - file: a file path (string or path-like object)
# - log_folder: a string for the folder name (default 'logs')

# Create a temporary file path to use
test_file = '/tmp/test_script.py'

ARGS = {
    'file': test_file,
    'log_folder': 'logs'
}
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 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
get_action Ran
ran on a synthesized input
vanderschaarlab/neurallaplacecontrol/mppi_with_model.py
code served (permissive licence) · get_code("cdf86004b2feba7a")
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

torch.manual_seed(0)

# Construct arguments for get_action
action_buffer = torch.randn(4, 3)  # Small 2D tensor (batch-like dimension 4, feature dimension 3)
action = torch.randn(3)  # Single action vector matching feature dimension
action_delay = 1  # Integer delay value

ARGS = {
    "action_buffer": action_buffer,
    "action": action,
    "action_delay": action_delay,
}
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
get_action_with_encode_obs_time Ran
ran on a synthesized input
vanderschaarlab/neurallaplacecontrol/mppi_dataset_collector.py
code served (permissive licence) · get_code("d6188eb95247b9a9")
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

torch.manual_seed(0)

# Analyze function usage:
# - action_buffer: used with torch.roll (needs to be a tensor), indexed with [-1, :nu] and [-(action_delay + 1), :nu]
# - action: assigned to action_buffer[-1, :nu], so should match that shape
# - action_delay: used as -(action_delay + 1) for indexing
# - nu: used for slicing action_buffer

# Create small tensors with distinct dimensions
nu = 3  # control dimension
action_delay = 2
batch_size = 5  # buffer depth - needs to be > action_delay + 1

# action_buffer: shape (batch_size, nu + padding)
# We need nu for control, and nu for padding (the zeros assignment)
action_buffer = torch.randn(batch_size, 2 * nu)

# action: shape (nu,) to match action_buffer[-1, :nu]
action = torch.randn(nu)

ARGS = {
    "action_buffer": action_buffer,
    "action": action,
    "action_delay": action_delay,
    "nu": nu
}
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
get_delta_t_rnn_model Ran
ran on a synthesized input
vanderschaarlab/neurallaplacecontrol/train_utils.py
code served (permissive licence) · get_code("70bbd381984aa87a")
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 DeltaTRNN. 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 config class with the required attributes
class MinimalConfig:
    def __init__(self):
        self.rnn_hidden_units = 64
        self.encode_obs_time = False
        self.normalize = True
        self.normalize_time = False

# Set up dimensions
state_dim = 4
action_dim = 3

# Create small tensors for statistics
state_mean = torch.zeros(state_dim)
state_std = torch.ones(state_dim)
action_mean = torch.zeros(action_dim)
action_std = torch.ones(action_dim)

# Create config
config = MinimalConfig()

ARGS = {
    'state_dim': state_dim,
    'action_dim': action_dim,
    'state_mean': state_mean,
    'action_mean': action_mean,
    'state_std': state_std,
    'action_std': action_std,
    'config': config
}
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 DeltaTRNN — a DeltaTRNN output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DeltaTRNN 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_rnn_model Ran
ran on a synthesized input
vanderschaarlab/neurallaplacecontrol/train_utils.py
code served (permissive licence) · get_code("cf12e790ed80faaf")
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 RNN. 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 config class with required attributes
class Config:
    def __init__(self):
        self.rnn_hidden_units = 64
        self.encode_obs_time = False
        self.normalize = False

# Set up dimensions
state_dim = 4
action_dim = 3

# Create small tensors for statistics (used in normalization)
state_mean = torch.zeros(state_dim, dtype=torch.float32)
state_std = torch.ones(state_dim, dtype=torch.float32)
action_mean = torch.zeros(action_dim, dtype=torch.float32)
action_std = torch.ones(action_dim, dtype=torch.float32)

# Create config instance
config = Config()

# Construct ARGS dict with all parameters for get_rnn_model
ARGS = {
    'state_dim': state_dim,
    'action_dim': action_dim,
    'state_mean': state_mean,
    'action_mean': action_mean,
    'state_std': state_std,
    'action_std': action_std,
    'config': config
}
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 RNN — a RNN output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RNN 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
is_float Ran
ran, violated the contract we drafted
samholt/neurallaplacecontrol/process_results/process_logs.py
code served (permissive licence) · get_code("e2448496ae21bf51")
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 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_float takes a single parameter 'element' and tries to convert it to float
# It returns True if successful, False if ValueError is raised
# We need to provide a value for 'element' that can be tested

ARGS = {
    "element": "3.14"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
load_replay_buffer Ran
ran on a synthesized input
vanderschaarlab/neurallaplacecontrol/w_nl.py
code served (permissive licence) · get_code("33f9df36e2fc0cfb")
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 numpy as np
import tempfile
import os

# Create a temporary numpy file with a valid replay buffer dictionary
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "replay_buffer.npy")

# Create a minimal offline dataset structure
offline_dataset = {
    "observations": np.array([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]]),
    "actions": np.array([[0.1, 0.2], [0.3, 0.4]]),
    "rewards": np.array([1.0, 2.0]),
    "terminals": np.array([False, True]),
    "next_observations": np.array([[5.0, 6.0, 7.0, 8.0], [9.0, 10.0, 11.0, 12.0]])
}

# Save the dataset to a numpy file
np.save(temp_file, offline_dataset)

ARGS = {
    "fn": temp_file
}
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 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
mean_confidence_interval Ran
ran; our contract draft was wrong, not the code
samholt/neurallaplacecontrol/process_results/process_logs.py
code served (permissive licence) · get_code("c23c1dcf5ce07300")
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, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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
import scipy.stats

# Construct valid arguments for mean_confidence_interval
# The function takes data (array-like) and confidence (float between 0 and 1)

# Create sample data - a list of numeric values
data = [1.2, 1.5, 1.3, 1.4, 1.6, 1.5, 1.4, 1.3, 1.5, 1.4]

# Standard confidence level
confidence = 0.95

ARGS = {
    "data": data,
    "confidence": confidence
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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
pendulum_dynamics_dt_delay Ran
ran on a synthesized input
vanderschaarlab/neurallaplacecontrol/oracle.py
code served (permissive licence) · get_code("b1120c7a27bffa9a")
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, 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

torch.manual_seed(0)

# Construct tensors for the function arguments
# state: shape (batch_size, state_dim) where state_dim is 2 or 3
# For this setup, using state_dim=3 (costh, sinth, thdot)
batch_size = 2
state = torch.randn(batch_size, 3)

# perturbed_action: shape (batch_size, time_steps, action_dim)
# The function indexes with [:, -(delay + 1), :nu]
# So we need enough time steps to accommodate the delay
time_steps = 5
action_dim = 1
nu = 1
delay = 2
perturbed_action = torch.randn(batch_size, time_steps, action_dim)

# ts: timestep(s), can be 1D or 2D
# Function reshapes it to (-1, 1), so passing 1D tensor
ts = torch.tensor([0.01], dtype=torch.float32)

# Other parameters with defaults
ACTION_LOW = -2.0
ACTION_HIGH = 2.0
friction = False

ARGS = {
    'state': state,
    'perturbed_action': perturbed_action,
    'ts': ts,
    'delay': delay,
    'ACTION_LOW': ACTION_LOW,
    'ACTION_HIGH': ACTION_HIGH,
    'nu': nu,
    'friction': friction
}
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, 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
string_to_float_dict Ran
ran; our contract draft was wrong, not the code
samholt/neurallaplacecontrol/process_results/process_logs.py
code served (permissive licence) · get_code("ab950ea1b98403e2")
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 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)

# Analyze the function: string_to_float_dict takes a dict 'd' as parameter
# It iterates over k, v pairs and converts values to float if they are floats
# The function uses: d.items(), float(v), is_float(v)

# Create a test dictionary with mixed types
test_dict = {
    "a": "3.14",
    "b": "not_a_number",
    "c": 42,
    "d": "2.71828",
    "e": "hello",
    "f": 1.5
}

ARGS = {
    "d": test_dict
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
create_LatentODE_model Not yet run
unverified (harvested, no recorded run)
vanderschaarlab/neurallaplacecontrol/baseline_models/latent_ode_lib/create_latent_ode_model.py
code served (permissive licence) · get_code("9b9f619dc624b573")
How it ran

no invocation recorded.

create_LatentODE_model_direct Not yet run
unverified (harvested, no recorded run)
vanderschaarlab/neurallaplacecontrol/baseline_models/latent_ode_lib/create_latent_ode_model.py
code served (permissive licence) · get_code("8b5c0856989359e1")
How it ran

no invocation recorded.

parse_args Not yet run
unverified (harvested, no recorded run)
vanderschaarlab/neurallaplacecontrol/config.py
code served (permissive licence) · get_code("c03dbd5716bc8def")
How it ran

no invocation recorded.

setup_logger Not yet run
unverified (harvested, no recorded run)
vanderschaarlab/neurallaplacecontrol/overlay.py
code served (permissive licence) · get_code("3728451722b589af")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 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, 1 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.

FunctionAlso present inEarliest-datedOthers named
create_classifier 3 other papers
2 dated earlier, 1 later
1907.03907 (2019-07)
Latent ODEs for Irregularly-Sampled Time Series
2206.04843 (2022-06)
Neural Laplace: Learning diverse classes of differential equations in the Laplace domain
2602.03215 (2026-02)
Latent Neural Ordinary Differential Equations for Model-Informed Precision Dosing: Overcoming Structural Assumptions in Pharmacokinetics

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

Many real-world offline reinforcement learning (RL) problems involve continuous-time environments with delays. Such environments are characterized by two distinctive features: firstly, the state x(t) is observed at irregular time intervals, and secondly, the current action a(t) only affects the future state x(t + g) with an unknown delay g > 0. A prime example of such an environment is satellite control where the communication link between earth and a satellite causes irregular observations and delays. Existing offline RL algorithms have achieved success in environments with irregularly observed states in time or known delays. However, environments involving both irregular observations in time and unknown delays remains an open and challenging problem. To this end, we propose Neural Laplace Control, a continuous-time model-based offline RL method that combines a Neural Laplace dynamics model with a model predictive control (MPC) planner--and is able to learn from an offline dataset sampled with irregular time intervals from an environment that has a inherent unknown constant delay. We show experimentally on continuous-time delayed environments it is able to achieve near expert policy performance.

For agents

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

get_harvested_code_for_paper("2302.12604")
get_code_for_paper("2302.12604")
have("2302.12604")

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

curl https://syntology.ai/api/ran/2302.12604.json

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

[![Syntology run record](https://syntology.ai/api/ran/2302.12604.svg)](https://syntology.ai/paper/2302.12604)

Connect an agent — have() is free.