SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.17213 · NeurIPS · 2022

Deep Gaussian Process-based Multi-fidelity Bayesian Optimization for Simulated Chemical Reactors

Tom Savage, Nausheen Basha, Antonio Del, Rio Chanona, Omar Matar

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 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 10 have no contract verdict; 0 failures of our instrument; 2 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/2210.17213.json · badge: /api/ran/2210.17213.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
optimal-pse-lab/pulsed-reactor-optimisation community (classifier label)
classifier basis not recorded on the link
10 ran · 2 no recorded run
FunctionStatusWhere it lives
angular_distance Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/mesh_generation/coil_both.py
code served (permissive licence) · get_code("8c19f888b7d8287f")
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 ArrayImpl, 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 jax.numpy as jnp
import numpy as np

# Analyze the function:
# - x, y, c are used in arithmetic operations
# - The function computes: jnp.abs((x - y + c) % (c * 2) - c)
# - All parameters should be numeric (scalars or arrays compatible with jnp)

# Create small numeric values for testing
x = jnp.array(1.5)
y = jnp.array(0.5)
c = jnp.array(np.pi)

ARGS = {
    "x": x,
    "y": y,
    "c": c,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
exp_design_hf Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/utils_gp.py
code served (permissive licence) · get_code("f914bb034fbe46c4")
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 ArrayImpl, 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 jax.numpy as jnp
import jax
from functools import partial

# Create a mock posterior callable that returns a distribution constructor
class MockDistribution:
    def __init__(self, mean, cov):
        self._mean = mean
        self._cov = cov
    
    def mean(self):
        return self._mean
    
    def covariance(self):
        return self._cov

class MockPosterior:
    def __call__(self, learned_params, D):
        # Returns a function that takes inputs and returns a distribution
        def posterior_fn(inputs):
            batch_size = inputs.shape[0]
            mean = jnp.ones((batch_size, 1))
            cov = jnp.eye(1) * 0.5
            return MockDistribution(mean, cov)
        return posterior_fn

class MockLikelihood:
    def __call__(self, learned_params, latent_distribution):
        # Takes latent distribution and returns predictive distribution
        latent_mean = latent_distribution.mean()
        latent_cov = latent_distribution.covariance()
        predictive_mean = latent_mean * 2.0
        predictive_cov = latent_cov * 1.5
        return MockDistribution(predictive_mean, predictive_cov)

# Construct the gp dictionary
gp = {
    'posterior': MockPosterior(),
    'learned_params': {'scale': 1.0, 'noise': 0.1},
    'D': 4,
    'likelihood': MockLikelihood()
}

# Construct x input
x = jnp.array([0.5, 0.3, 0.2, 0.1])

ARGS = {
    'x': x,
    'gp': gp
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
format_data Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/utils.py
code served (permissive licence) · get_code("42c4b8839b4ec7a2")
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 jax.numpy as jnp

# Construct a valid data dict that format_data expects
data = {
    'data': [
        {
            'id': 'sample_1',
            'x': {'feature_0': 1.0, 'feature_1': 2.0, 'feature_2': 3.0, 'feature_3': 4.0},
            'obj': 5.5,
            'cost': 10.2
        },
        {
            'id': 'sample_2',
            'x': {'feature_0': 2.0, 'feature_1': 3.0, 'feature_2': 4.0, 'feature_3': 5.0},
            'obj': 6.5,
            'cost': 11.2
        },
        {
            'id': 'running',
            'x': {'feature_0': 9.0, 'feature_1': 9.0, 'feature_2': 9.0, 'feature_3': 9.0},
            'obj': 99.0,
            'cost': 99.0
        }
    ]
}

ARGS = {
    'data': data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — 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
mean_std Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/utils.py
code served (permissive licence) · get_code("721211fed75787ee")
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, ndim 2, 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

# The function mean_std takes a single parameter 'vec'
# It calls np.mean(vec, axis=0) and np.std(vec, axis=0)
# This means vec should be a 2D or higher dimensional numpy array
# where axis=0 is the batch/sample dimension

# Create a small 2D numpy array (batch_size=2, features=4)
vec = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
], dtype=np.float32)

ARGS = {
    "vec": vec
}
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, ndim 2, 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
normalise Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/utils.py
code served (permissive licence) · get_code("0987413217de1d2f")
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)

# The function normalise takes three parameters: vec, mean, std
# It performs element-wise operations: (vec - mean) / std
# All three should be tensors or array-like that support subtraction and division

vec = torch.randn(2, 4)
mean = torch.randn(2, 4)
std = torch.randn(2, 4)

ARGS = {
    "vec": vec,
    "mean": mean,
    "std": std,
}
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
rotate_x Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/mesh_generation/coil_basic.py
code served (permissive licence) · get_code("2c8388bebe784bb1")
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, ndim 2, 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

# Construct arguments for rotate_x function
# Parameters: x0, y0, z0, r_x

x0 = np.array([1.0, 2.0, 3.0, 4.0])
y0 = np.array([5.0, 6.0, 7.0, 8.0])
z0 = np.array([9.0, 10.0, 11.0, 12.0])
r_x = np.pi / 4  # rotation angle in radians

ARGS = {
    "x0": x0,
    "y0": y0,
    "z0": z0,
    "r_x": r_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 tuple, ndim 2, 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
rotate_x_simple Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/mesh_generation/coil_experimental.py
code served (permissive licence) · get_code("f4ea90a0269cb71d")
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, ndim 2, 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

# Construct arguments for rotate_x_simple
x0 = [1.0, 2.0, 3.0]
y0 = np.array([0.0, 1.0, 2.0])
z0 = np.array([0.5, 1.5, 2.5])
r_x = np.pi / 4  # 45 degrees rotation angle

ARGS = {
    'x0': x0,
    'y0': y0,
    'z0': z0,
    'r_x': r_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 tuple, ndim 2, 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
rotate_y Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/mesh_generation/coil_basic.py
code served (permissive licence) · get_code("a693e601e8768688")
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, ndim 2, 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

# Construct arguments for rotate_y function
# The function performs a rotation around the y-axis on 3D coordinates

x0 = np.array([1.0, 2.0, 3.0, 4.0])
y0 = np.array([5.0, 6.0, 7.0, 8.0])
z0 = np.array([9.0, 10.0, 11.0, 12.0])
r_y = np.pi / 4  # 45 degrees rotation around y-axis

ARGS = {
    'x0': x0,
    'y0': y0,
    'z0': z0,
    'r_y': r_y
}
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, ndim 2, 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
rotate_z Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/mesh_generation/coil_basic.py
code served (permissive licence) · get_code("1640622ed1bb755a")
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, ndim 2, 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

# The rotate_z function takes x, y, z coordinate arrays and a rotation angle r_z
# It performs a 2D rotation around the z-axis using rotation matrices

# Create small coordinate arrays with distinct sizes for clarity
x = np.array([1.0, 2.0, 3.0])
y = np.array([4.0, 5.0, 6.0])
z = np.array([7.0, 8.0, 9.0])

# Rotation angle in radians
r_z = np.pi / 4  # 45 degrees

ARGS = {
    'x': x,
    'y': y,
    'z': z,
    'r_z': r_z
}
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, ndim 2, 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
rotate_z Ran
ran on a synthesized input
optimal-pse-lab/pulsed-reactor-optimisation/mesh_generation/coil_both.py
code served (permissive licence) · get_code("ea6852881d861795")
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, ndim 2, 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

# Set random seed for reproducibility
np.random.seed(0)

# Create sample data for rotation
x0 = np.array([1.0, 2.0, 3.0, 4.0])
y0 = np.array([5.0, 6.0, 7.0, 8.0])
z0 = np.array([9.0, 10.0, 11.0, 12.0])
r_z = np.pi / 4  # 45 degrees rotation angle

ARGS = {
    "x0": x0,
    "y0": y0,
    "z0": z0,
    "r_z": r_z
}
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, ndim 2, 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
compute_mutual_info Not yet run
unverified (harvested, no recorded run)
optimal-pse-lab/pulsed-reactor-optimisation/utils_gp.py
code served (permissive licence) · get_code("ad1dbf3214875771")
How it ran

no invocation recorded.

exp_design_function Not yet run
unverified (harvested, no recorded run)
optimal-pse-lab/pulsed-reactor-optimisation/utils_gp.py
code served (permissive licence) · get_code("2d80a930a7909e37")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 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 12 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

New manufacturing techniques such as 3D printing have recently enabled the creation of previously infeasible chemical reactor designs. Optimizing the geometry of the next generation of chemical reactors is important to understand the underlying physics and to ensure reactor feasibility in the real world. This optimization problem is computationally expensive, nonlinear, and derivative-free making it challenging to solve. In this work, we apply deep Gaussian processes (DGPs) to model multi-fidelity coiled-tube reactor simulations in a Bayesian optimization setting. By applying a multi-fidelity Bayesian optimization method, the search space of reactor geometries is explored through an amalgam of different fidelity simulations which are chosen based on prediction uncertainty and simulation cost, maximizing the use of computational budget. The use of DGPs provides an endto-end model for five discrete mesh fidelities, enabling less computational effort to gain good solutions during optimization. The accuracy of simulations for these five fidelities is determined against experimental data obtained from a 3D printed reactor configuration, providing insights into appropriate hyper-parameters. We hope this work provides interesting insight into the practical use of DGP-based multi-fidelity Bayesian optimization for engineering discovery. 2022 NeurIPS Workshop on Gaussian Processes, Spatiotemporal Modeling, and Decision-making Systems.

For agents

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

get_harvested_code_for_paper("2210.17213")
get_code_for_paper("2210.17213")
have("2210.17213")

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

curl https://syntology.ai/api/ran/2210.17213.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.17213.svg)](https://syntology.ai/paper/2210.17213)

Connect an agent — have() is free.