SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.16795 · ICML · 2024

In-Context Freeze-Thaw Bayesian Optimization for Hyperparameter Optimization

Frank Hutter, Neeratyoy Mallik, Herilalaina Rakotoarison, Steven Adriaensen, Samir Garibov, Edward Bergman

arXiv · PDF · Open in the Atlas

Code that ran

3 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 1 failure of our instrument; 8 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/2404.16795.json · badge: /api/ran/2404.16795.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
automl/ifbo canonical (classifier label)
classifier basis: classifier_2025
3 ran · 8 no recorded run
FunctionStatusWhere it lives
BarDistribution Ran
ran on a synthesized input · constructed an object; no result was computed
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("ae93b38649dac0ed")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, BarDistribution: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from torch import nn

torch.manual_seed(0)

# Create borders tensor - must be sorted and span a range
# Using values from 0 to 10 with 5 bins
borders = torch.tensor([0.0, 2.0, 4.0, 6.0, 8.0, 10.0])

# Smoothing parameter
smoothing = 0.1

# ignore_nan_targets parameter
ignore_nan_targets = True

ARGS = {
    'borders': borders,
    'smoothing': smoothing,
    'ignore_nan_targets': ignore_nan_targets,
}
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 BarDistribution — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a BarDistribution 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
Curve Ran
ran on a synthesized input · constructed an object; no result was computed
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("bd43d10d65fb49ec")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, Curve: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create tiny tensors for the Curve dataclass
# hyperparameters: should be <= 10 dimensions, values in [0, 1]
hyperparameters = torch.rand(2, 4)  # batch_size=2, dim=4

# t: time steps in [0, 1]
t = torch.rand(2, 3)  # batch_size=2, num_timesteps=3

# y: performance values in [0, 1], optional
y = torch.rand(2, 3)  # batch_size=2, num_timesteps=3

ARGS = {
    "hyperparameters": hyperparameters,
    "t": t,
    "y": y,
}
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 Curve — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Curve 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
_resolve_model_path Ran
ran; our contract draft was wrong, not the code
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("d8a5f8cf9fcd07e2")
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 PosixPath. 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.

from pathlib import Path
import tempfile
import shutil

# Create a temporary directory for testing
temp_dir = tempfile.mkdtemp()

# Test case 1: None (will use default)
target_path_none = None

# Test case 2: Valid directory path
target_path_valid = Path(temp_dir) / "models"
target_path_valid.mkdir(parents=True, exist_ok=True)

ARGS = {
    "target_path": target_path_valid
}
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 PosixPath — a PosixPath output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PosixPath 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
FILENAME Not yet run
unverified (harvested, no recorded run)
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("864f165e47883adb")
How it ran

no invocation recorded.

FILE_URL Not yet run
unverified (harvested, no recorded run)
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("b9bfe00fed47e0b8")
How it ran

no invocation recorded.

FTPFN Not yet run
unverified (harvested, no recorded run)
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("eaec9f9ba511916f")
How it ran

no invocation recorded.

PredictionResult Not yet run
unverified (harvested, no recorded run)
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("6e2a88a13e2d4ad3")
How it ran

no invocation recorded.

WEIGHTS_FINAL_NAME Not yet run
unverified (harvested, no recorded run)
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("593bf795eb439666")
How it ran

no invocation recorded.

_fetch_response_with_repeats Not yet run
unverified (harvested, no recorded run)
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("81b495c548baefa9")
How it ran

no invocation recorded.

download_and_decompress Not yet run
unverified (harvested, no recorded run)
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("539085ca4dfb5e3d")
How it ran

no invocation recorded.

tokenize Not yet run
unverified (harvested, no recorded run)
automl/ifbo/ifbo/surrogate.py
code served (permissive licence) · get_code("d2413f6e81b6146b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 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 9 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
FILENAME
brick of this paper's assembly: helper
1 other paper
1 later
2510.21379 (2025-10)
Cost-Sensitive Freeze-thaw Bayesian Optimization for Efficient Hyperparameter Tuning
FILE_URL
brick of this paper's assembly: helper
1 other paper
1 later
2510.21379 (2025-10)
Cost-Sensitive Freeze-thaw Bayesian Optimization for Efficient Hyperparameter Tuning

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

With the increasing computational costs associated with deep learning, automated hyperparameter optimization methods, strongly relying on black-box Bayesian optimization (BO), face limitations. Freeze-thaw BO offers a promising grey-box alternative, strategically allocating scarce resources incrementally to different configurations. However, the frequent surrogate model updates inherent to this approach pose challenges for existing methods, requiring retraining or fine-tuning their neural network surrogates online, introducing overhead, instability, and hyper-hyperparameters. In this work, we propose FT-PFN, a novel surrogate for Freezethaw style BO. FT-PFN is a prior-data fitted network (PFN) that leverages the transformers' in-context learning ability to efficiently and reliably do Bayesian learning curve extrapolation in a single forward pass. Our empirical analysis across three benchmark suites shows that the predictions made by FT-PFN are more accurate and 10-100 times faster than those of the deep Gaussian process and deep ensemble surrogates used in previous work. Furthermore, we show that, when combined with our novel acquisition mechanism (MFPI-random), the resulting in-context freeze-thaw BO method (ifBO), yields new stateof-the-art performance in the same three families of deep learning HPO benchmarks considered in prior work.

For agents

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

get_harvested_code_for_paper("2404.16795")
get_code_for_paper("2404.16795")
have("2404.16795")

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

curl https://syntology.ai/api/ran/2404.16795.json

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

[![Syntology run record](https://syntology.ai/api/ran/2404.16795.svg)](https://syntology.ai/paper/2404.16795)

Connect an agent — have() is free.