SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2412.12918 · 2024

BOIDS: High-dimensional Bayesian Optimization via Incumbent-guided Direction Lines and Subspace Embeddings

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 8 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/2412.12918.json · badge: /api/ran/2412.12918.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
LamNgo1/boids canonical (classifier label)
classifier basis: census
9 ran · 0 no recorded run
FunctionStatusWhere it lives
from_unit_cube Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LamNgo1/boids/baxus_embedding/utils.py
code served (permissive licence) · get_code("3ec3344b3381a5d8")
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 ndarray, 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 numpy as np

# Create small test arrays
# x: points in unit cube space (batch_size=2, dimensions=4)
x = np.random.rand(2, 4).astype(np.float32)

# lower_bounds: lower bounds of full space (dimensions=4)
lower_bounds = np.array([-1.0, -2.0, -3.0, -4.0], dtype=np.float32)

# upper_bounds: upper bounds of full space (dimensions=4)
upper_bounds = np.array([1.0, 2.0, 3.0, 4.0], dtype=np.float32)

ARGS = {
    "x": x,
    "lower_bounds": lower_bounds,
    "upper_bounds": upper_bounds,
}
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 ndarray, 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
get_acquisition_function Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LamNgo1/boids/igline_learning/multiobjective_acquisition.py
code served (permissive licence) · get_code("35d0c3d16a599523")
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 function. 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
import gpytorch
from botorch.models import SingleTaskGP
from gpytorch.likelihoods import GaussianLikelihood

torch.manual_seed(0)
np.random.seed(0)

# Create minimal training data
train_x = torch.randn(5, 3)  # 5 points, 3 features
train_y = torch.randn(5, 1)  # 5 targets

# Create a SingleTaskGP model
gp = SingleTaskGP(train_x, train_y)

# Define parameters
acq = 'ei'  # Can be 'ts', 'ei', 'logei', or 'ucb'
observed_mu = np.array([0.5, 0.3])
observed_sigma = np.array([0.1, 0.2])
dtype = torch.float32
device = torch.device('cpu')
max_cholesky_size = 2000

ARGS = {
    'acq': acq,
    'gp': gp,
    'observed_mu': observed_mu,
    'observed_sigma': observed_sigma,
    'dtype': dtype,
    'device': device,
    'max_cholesky_size': max_cholesky_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
botorch, gpytorch, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
initializer_factory Ran
ran; our contract draft was wrong, not the code
LamNgo1/boids/surrogate/gp_utils.py
code served (permissive licence) · get_code("20169e622fc160cb")
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 function. 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
from typing import Dict

torch.manual_seed(0)

# Create a minimal stand-in GP model class that has an initialize method
class MockGP(nn.Module):
    def __init__(self):
        super().__init__()
        self.mean_module = None
        self.covar_module = None
        self.likelihood = None
    
    def initialize(self, **kwargs):
        """Mock initialize method that accepts hyperparameters"""
        for key, value in kwargs.items():
            setattr(self, key, value)

# Create a sample hyperparameter configuration
hyperparameter_configuration = {
    "mean_constant": 0.5,
    "outputscale": 1.2,
    "lengthscale": 0.8,
    "noise": 0.1
}

ARGS = {
    "hyperparameter_configuration": hyperparameter_configuration
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
join_data Ran
ran; our contract draft was wrong, not the code
LamNgo1/boids/baxus_embedding/data_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from LeoIV/BAxUS, is pointer only (licence: NOASSERTION) · get_code("df14ac18e98ce26b")
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 LeoIV/BAxUS, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

mle_optimization Ran
ran; our contract draft was wrong, not the code
LamNgo1/boids/surrogate/gp_utils.py
code served (permissive licence) · get_code("a919c6c448991d83")
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
import torch.nn as nn
from gpytorch.models import ExactGP
from gpytorch.means import ConstantMean
from gpytorch.kernels import ScaleKernel, RBFKernel
from gpytorch.distributions import MultivariateNormal
from gpytorch.likelihoods import GaussianLikelihood
from torch.distributions import MultivariateNormal as TorchMVN

torch.manual_seed(0)

# Create a minimal GP model that satisfies the function's requirements
class MinimalGP(ExactGP):
    def __init__(self, train_x, train_y, likelihood):
        super().__init__(train_x, train_y, likelihood)
        self.mean_module = ConstantMean()
        self.covar_module = ScaleKernel(RBFKernel())
    
    def forward(self, x):
        mean_x = self.mean_module(x)
        covar_x = self.covar_module(x)
        return MultivariateNormal(mean_x, covar_x)

# Create training data
train_x = torch.randn(2, 4)  # batch_size=2, input_dim=4
train_y = torch.randn(2)     # batch_size=2

# Create likelihood and model
likelihood = GaussianLikelihood()
model = MinimalGP(train_x, train_y, likelihood)

# Define an initializer function that takes a model and initializes it
def initializer(gp_model):
    # Simple initialization: set hyperparameters to some values
    gp_model.likelihood.noise = torch.tensor(0.1)

ARGS = {
    'initializer': initializer,
    'model': model,
    'num_steps': 2,
    'train_x': train_x,
    'train_y': train_y,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gpytorch, 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
pick_best_from_configurations Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LamNgo1/boids/surrogate/gp_utils.py
code served (permissive licence) · get_code("15daf13f27b6fa56")
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 list, 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 torch
import torch.nn as nn
from gpytorch.models import ExactGP
from gpytorch.means import ConstantMean
from gpytorch.kernels import ScaleKernel, RBFKernel
from gpytorch.distributions import MultivariateNormal
from gpytorch.likelihoods import GaussianLikelihood

torch.manual_seed(0)

# Create a minimal GP model class
class SimpleGP(ExactGP):
    def __init__(self, train_x, train_y, likelihood):
        super().__init__(train_x, train_y, likelihood)
        self.mean_module = ConstantMean()
        self.covar_module = ScaleKernel(RBFKernel())
    
    def forward(self, x):
        mean_x = self.mean_module(x)
        covar_x = self.covar_module(x)
        return MultivariateNormal(mean_x, covar_x)

# Create tiny training data
train_x = torch.randn(2, 4)  # 2 samples, 4 dimensions
train_y = torch.randn(2)     # 2 targets

# Create likelihood and model
likelihood = GaussianLikelihood()
model = SimpleGP(train_x, train_y, likelihood)

# Create initializer functions that set hyperparameters
def initializer_1(gp_model):
    gp_model.mean_module.constant.data.fill_(0.5)
    gp_model.covar_module.base_kernel.lengthscale = 1.0

def initializer_2(gp_model):
    gp_model.mean_module.constant.data.fill_(1.0)
    gp_model.covar_module.base_kernel.lengthscale = 0.5

def initializer_3(gp_model):
    gp_model.mean_module.constant.data.fill_(0.1)
    gp_model.covar_module.base_kernel.lengthscale = 2.0

initializers = [initializer_1, initializer_2, initializer_3]

ARGS = {
    'initializers': initializers,
    'model': model,
    'train_x': train_x,
    'train_y': train_y,
    'n_best': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gpytorch, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list 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
right_pad_sequence Ran
ran, honoured the contract we drafted
LamNgo1/boids/baxus_embedding/data_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from LeoIV/BAxUS, is pointer only (licence: NOASSERTION) · get_code("a8eee1e25d7db974")
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 LeoIV/BAxUS, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

to_1_around_origin Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LamNgo1/boids/baxus_embedding/utils.py
code served (permissive licence) · get_code("f80a39b33f318b88")
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 ndarray, 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 numpy as np

# Create small test data
# x: batch of 2 points in 4 dimensions
x = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
], dtype=np.float32)

# lower_bounds: 4 dimensional lower bounds
lower_bounds = np.array([0.0, 0.0, 0.0, 0.0], dtype=np.float32)

# upper_bounds: 4 dimensional upper bounds
upper_bounds = np.array([10.0, 10.0, 10.0, 10.0], dtype=np.float32)

ARGS = {
    'x': x,
    'lower_bounds': lower_bounds,
    'upper_bounds': upper_bounds
}
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 ndarray, 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
to_unit_cube Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LamNgo1/boids/baxus_embedding/utils.py
code served (permissive licence) · get_code("fd482e1d259adbd4")
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 ndarray, 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 numpy as np

# Create small example data
np.random.seed(0)

# Points to scale: shape (2, 4) - 2 points in 4D space
x = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
], dtype=np.float32)

# Lower bounds: shape (4,) - one lower bound per dimension
lower_bounds = np.array([0.0, 1.0, 2.0, 3.0], dtype=np.float32)

# Upper bounds: shape (4,) - one upper bound per dimension
upper_bounds = np.array([10.0, 11.0, 12.0, 13.0], dtype=np.float32)

ARGS = {
    "x": x,
    "lower_bounds": lower_bounds,
    "upper_bounds": upper_bounds
}
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 ndarray, 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

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 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.

FunctionAlso present inEarliest-datedOthers named
join_data 1 other paper
1 later
2608.15073 (2026-08)
BOCoDe: Engineering-Centered Benchmarking for Bayesian Optimization
right_pad_sequence 1 other paper
1 later
2608.15073 (2026-08)
BOCoDe: Engineering-Centered Benchmarking for Bayesian Optimization

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

When it comes to expensive black-box optimization problems, Bayesian Optimization (BO) is a well-known and powerful solution. Many real-world applications involve a large number of dimensions, hence scaling BO to high dimension is of much interest. However, state-of-the-art high-dimensional BO methods still suffer from the curse of dimensionality, highlighting the need for further improvements. In this work, we introduce BOIDS, a novel high-dimensional BO algorithm that guides optimization by a sequence of one-dimensional direction lines using a novel tailored line-based optimization procedure. To improve the efficiency, we also propose an adaptive selection technique to identify most optimal lines for each round of line-based optimization. Additionally, we incorporate a subspace embedding technique for better scaling to high-dimensional spaces. We further provide theoretical analysis of our proposed method to analyze its convergence property. Our extensive experimental results show that BOIDS outperforms state-of-the-art baselines on various synthetic and real-world benchmark problems.

For agents

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

get_harvested_code_for_paper("2412.12918")
get_code_for_paper("2412.12918")
have("2412.12918")

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

curl https://syntology.ai/api/ran/2412.12918.json

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

[![Syntology run record](https://syntology.ai/api/ran/2412.12918.svg)](https://syntology.ai/paper/2412.12918)

Connect an agent — have() is free.