SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1911.12809 · 2019

Greed is Good: Exploration and Exploitation Trade-offs in Bayesian Optimisation

arXiv · PDF · Open in the Atlas

Code that ran

6 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; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 0 failures of our instrument; 3 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/1911.12809.json · badge: /api/ran/1911.12809.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
georgedeath/egreedy canonical (classifier label)
classifier basis: census
6 ran · 3 no recorded run
FunctionStatusWhere it lives
EI Ran
ran on a synthesized input
georgedeath/egreedy/egreedy/acquisition_functions/standard_acq_funcs_minimize.py
code served (permissive licence) · get_code("200d26a3d7ada429")
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 (4, 1) 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
from scipy.stats import norm

# Create small example data for the EI function
mu = np.array([1.5, 2.3, 0.8, 1.1])
sigma = np.array([0.5, 0.3, 0.0, 0.4])
y_best = 0.5

ARGS = {
    "mu": mu,
    "sigma": sigma,
    "y_best": y_best,
    "PHI": norm.cdf,
    "phi": norm.pdf,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 (4, 1) — shape (4, 1) 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
PFRandom Ran
ran on a synthesized input
georgedeath/egreedy/egreedy/acquisition_functions/egreedy_acq_funcs_minimize.py
code served (permissive licence) · get_code("b678c5f9f5359435")
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 1, dtype family float, shape (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 numpy as np

# Create a small Pareto front with N=5 locations in D=3 dimensional space
np.random.seed(0)
X = np.array([
    [0.1, 0.2, 0.3],
    [0.4, 0.1, 0.2],
    [0.2, 0.3, 0.1],
    [0.3, 0.4, 0.2],
    [0.15, 0.25, 0.35]
], dtype=np.float32)

ARGS = {
    "X": 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 ndarray, ndim 1, dtype family float, shape (3,) — shape (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
PI Ran
ran on a synthesized input
georgedeath/egreedy/egreedy/acquisition_functions/standard_acq_funcs_minimize.py
code served (permissive licence) · get_code("68a3c0db517ad373")
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 (4, 1) 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
from scipy.stats import norm

# Create sample data for the PI function
mu = np.array([1.5, 2.3, 0.8, 3.1])
sigma = np.array([0.5, 0.3, 0.0, 0.7])
y_best = 2.0

# phi is already the default norm.pdf from scipy.stats
phi = norm.pdf

ARGS = {
    'mu': mu,
    'sigma': sigma,
    'y_best': y_best,
    'phi': phi
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 (4, 1) — shape (4, 1) 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
UCB Ran
ran on a synthesized input
georgedeath/egreedy/egreedy/acquisition_functions/standard_acq_funcs_minimize.py
code served (permissive licence) · get_code("065426747a4fd65a")
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 1, dtype family float, shape (5,) 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

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

# Create small arrays for mu and sigma
N = 5  # number of locations
mu = np.array([0.5, 1.2, -0.3, 0.8, 1.5], dtype=np.float64)
sigma = np.array([0.1, 0.2, 0.15, 0.25, 0.12], dtype=np.float64)

# Bounds for the search space
lb = np.array([-1.0, -1.0, -1.0, -1.0, -1.0], dtype=np.float64)
ub = np.array([2.0, 2.0, 2.0, 2.0, 2.0], dtype=np.float64)

# Time step
t = 10

# Dimensionality
d = 3

# Beta calculation method
beta = 'theorem2'

# Size of design set (for theorem1)
sizeD = 1000

# Parameters for theorem2
a = 1.0
b = 1.0

# Parameter for theorem3
B = 10

ARGS = {
    'mu': mu,
    'sigma': sigma,
    'lb': lb,
    'ub': ub,
    't': t,
    'd': d,
    'beta': beta,
    'sizeD': sizeD,
    'a': a,
    'b': b,
    'B': B,
}
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 1, dtype family float, shape (5,) — shape (5,) 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
eFront Ran
ran on a synthesized input
georgedeath/egreedy/egreedy/acquisition_functions/egreedy_acq_funcs_minimize.py
code served (permissive licence) · get_code("8ceb86a31b462cf2")
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 1, dtype family float, shape (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

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

# Create sample data for the Pareto front
N = 5  # Number of points on the Pareto front
D = 4  # Dimensionality of decision space

# X: (N, D) array of locations in decision space
X = np.random.rand(N, D)

# mu: (N,) array of mean value predictions
mu = np.array([0.5, 0.3, 0.8, 0.2, 0.6])

# sigma: (N,) array of standard deviations
sigma = np.array([0.1, 0.15, 0.05, 0.12, 0.08])

# epsilon: probability of random selection
epsilon = 0.1

ARGS = {
    "X": X,
    "mu": mu,
    "sigma": sigma,
    "epsilon": epsilon
}
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 1, dtype family float, shape (4,) — shape (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
eRandom Ran
ran on a synthesized input
georgedeath/egreedy/egreedy/acquisition_functions/egreedy_acq_funcs_minimize.py
code served (permissive licence) · get_code("9c15aa08b0ecead0")
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 1, dtype family float, shape (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 numpy as np

np.random.seed(0)

# Create a small Pareto front with 5 points in 3D decision space
N = 5  # number of points on Pareto front
D = 3  # decision space dimensionality

X = np.array([
    [0.1, 0.2, 0.3],
    [0.4, 0.5, 0.6],
    [0.7, 0.8, 0.9],
    [0.2, 0.3, 0.4],
    [0.5, 0.6, 0.7]
])

mu = np.array([1.5, 2.3, 1.2, 3.1, 2.0])

sigma = np.array([0.1, 0.2, 0.15, 0.25, 0.18])

lb = np.array([0.0, 0.0, 0.0])

ub = np.array([1.0, 1.0, 1.0])

# Optional constraint function that checks if a point is within bounds
# (though the function already uses lb and ub for uniform sampling)
def cf(x):
    return np.all(x >= lb) and np.all(x <= ub)

epsilon = 0.3

ARGS = {
    "X": X,
    "mu": mu,
    "sigma": sigma,
    "lb": lb,
    "ub": ub,
    "cf": cf,
    "epsilon": epsilon,
}
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 1, dtype family float, shape (3,) — shape (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
make_1thing Not yet run
unverified (harvested, no recorded run)
georgedeath/egreedy/egreedy/test_problems/push_world.py
code served (permissive licence) · get_code("080e6f7ac4e38524")
How it ran

no invocation recorded.

make_base Not yet run
unverified (harvested, no recorded run)
georgedeath/egreedy/egreedy/test_problems/push_world.py
code served (permissive licence) · get_code("e5b288345afd0b4a")
How it ran

no invocation recorded.

simu_push_4D Not yet run
unverified (harvested, no recorded run)
georgedeath/egreedy/egreedy/test_problems/push_world.py
code served (permissive licence) · get_code("16ec59f1c3ac5771")
How it ran

no invocation recorded.

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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 8 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
simu_push_4D 1 other paper
1 later
2004.08349 (2020-04)
What do you Mean? The Role of the Mean Function in Bayesian Optimisation

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

The performance of acquisition functions for Bayesian optimisation to locate the global optimum of continuous functions is investigated in terms of the Pareto front between exploration and exploitation. We show that Expected Improvement (EI) and the Upper Confidence Bound (UCB) always select solutions to be expensively evaluated on the Pareto front, but Probability of Improvement is not guaranteed to do so and Weighted Expected Improvement does so only for a restricted range of weights. We introduce two novel $ε$-greedy acquisition functions. Extensive empirical evaluation of these together with random search, purely exploratory, and purely exploitative search on 10 benchmark problems in 1 to 10 dimensions shows that $ε$-greedy algorithms are generally at least as effective as conventional acquisition functions (e.g., EI and UCB), particularly with a limited budget. In higher dimensions $ε$-greedy approaches are shown to have improved performance over conventional approaches. These results are borne out on a real world computational fluid dynamics optimisation problem and a robotics active learning problem. Our analysis and experiments suggest that the most effective strategy, particularly in higher dimensions, is to be mostly greedy, occasionally selecting a random exploratory solution.

For agents

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

get_harvested_code_for_paper("1911.12809")
get_code_for_paper("1911.12809")
have("1911.12809")

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

curl https://syntology.ai/api/ran/1911.12809.json

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

[![Syntology run record](https://syntology.ai/api/ran/1911.12809.svg)](https://syntology.ai/paper/1911.12809)

Connect an agent — have() is free.