SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.07034 · CVPR · 2022

Active Learning by Feature Mixing

Ehsan Abbasnejad, Javen Shi, Damien Teney, Anton Van Den Hengel, Reza Haffari, Amin Parvaneh

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (5 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 0 failures of our instrument; 1 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/2203.07034.json · badge: /api/ran/2203.07034.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
dsba-lab/openal community
Papers with Code archive
6 ran · 0 no recorded run
aminparvaneh/alpha_mix_active_learning marked official in Papers with Code
Papers with Code archive
1 ran · 1 no recorded run
FunctionStatusWhere it lives
AlphaMixSampling Ran
ran on a synthesized input · constructed an object; no result was computed
dsba-lab/openal/query_strategies/featmix_sampling.py
code served (permissive licence) · get_code("e75f4ac8794a8b1c")
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, AlphaMixSampling: 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
import torch.nn as nn
import numpy as np
from torch.utils.data import Dataset

torch.manual_seed(0)

# Create a minimal model that the Strategy class can use
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc1 = nn.Linear(4, 8)
        self.fc2 = nn.Linear(8, 3)
    
    def forward(self, x):
        x = self.fc1(x)
        x = torch.relu(x)
        x = self.fc2(x)
        return x

# Create a minimal dataset
class SimpleDataset(Dataset):
    def __init__(self, size=10, input_dim=4, num_classes=3):
        self.size = size
        self.input_dim = input_dim
        self.num_classes = num_classes
        self.data = torch.randn(size, input_dim)
        self.labels = torch.randint(0, num_classes, (size,))
    
    def __len__(self):
        return self.size
    
    def __getitem__(self, idx):
        return self.data[idx], self.labels[idx]

# Create minimal transform
class SimpleTransform:
    def __call__(self, x):
        return x

# Set up arguments
model = SimpleModel()
n_query = 2
dataset = SimpleDataset(size=10, input_dim=4, num_classes=3)
transform = SimpleTransform()
is_labeled = np.array([True, True, False, False, False, False, False, False, False, False])
batch_size = 2
num_workers = 0
sampler_name = 'default'
steps_per_epoch = 0
n_subset = 0
is_openset = False
is_unlabeled = None
is_ood = None
id_classes = None
use_diverse = False

ARGS = {
    'model': model,
    'n_query': n_query,
    'dataset': dataset,
    'transform': transform,
    'is_labeled': is_labeled,
    'batch_size': batch_size,
    'num_workers': num_workers,
    'sampler_name': sampler_name,
    'steps_per_epoch': steps_per_epoch,
    'n_subset': n_subset,
    'is_openset': is_openset,
    'is_unlabeled': is_unlabeled,
    'is_ood': is_ood,
    'id_classes': id_classes,
    'use_diverse': use_diverse,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type AlphaMixSampling — 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 AlphaMixSampling 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
Strategy Ran
ran on a synthesized input · constructed an object; no result was computed
dsba-lab/openal/query_strategies/featmix_sampling.py
code served (permissive licence) · get_code("bf475d60bd681390")
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, Strategy: 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
import torch.nn as nn
import numpy as np
from torch.utils.data import Dataset

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc1 = nn.Linear(4, 8)
        self.fc2 = nn.Linear(8, 3)
    
    def forward(self, x, return_embed=False):
        embed = self.fc1(x)
        logits = self.fc2(embed)
        if return_embed:
            return {'logits': logits, 'embed': embed}
        return logits

class SimpleDataset(Dataset):
    def __init__(self, size=20):
        self.size = size
        self.data = torch.randn(size, 4)
        self.targets = np.random.randint(0, 3, size)
    
    def __len__(self):
        return self.size
    
    def __getitem__(self, idx):
        return self.data[idx], self.targets[idx]

model = SimpleModel()
dataset = SimpleDataset(size=20)
n_query = 3
batch_size = 2
num_workers = 0
is_labeled = np.array([True, True, False, False, False, False, False, False, False, False, 
                       False, False, False, False, False, False, False, False, False, False])
is_unlabeled = np.array([False, False, True, True, True, True, True, True, True, True,
                         True, True, True, True, True, True, True, True, True, True])
is_ood = np.array([False] * 20)
id_classes = np.array([0, 1, 2])

def simple_transform(x):
    return x

ARGS = {
    'model': model,
    'n_query': n_query,
    'dataset': dataset,
    'transform': simple_transform,
    'is_labeled': is_labeled,
    'batch_size': batch_size,
    'num_workers': num_workers,
    'sampler_name': 'RandomSampler',
    'steps_per_epoch': 0,
    'n_subset': 0,
    'is_openset': True,
    'is_unlabeled': is_unlabeled,
    'is_ood': is_ood,
    'id_classes': id_classes,
    'use_diverse': False,
    'accelerator': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Strategy — 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 Strategy 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
Strategy Ran
ran on a synthesized input
aminparvaneh/alpha_mix_active_learning/query_strategies/alpha_mix_sampling.py
pointer only (licence: NONE) · get_code("2661daeb22d41bb0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, aminparvaneh/alpha_mix_active_learning 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.

SubsetSequentialSampler Ran
ran on a synthesized input · constructed an object; no result was computed
dsba-lab/openal/query_strategies/featmix_sampling.py
code served (permissive licence) · get_code("c15a37c34d489f09")
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, SubsetSequentialSampler: 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 numpy as np
from torch.utils.data import Sampler

class SubsetSequentialSampler(Sampler):

    def __init__(self, indices: np.ndarray):
        self.indices = indices

    def __iter__(self):
        return iter(self.indices)

    def __len__(self):
        return len(self.indices)


# Create valid arguments for SubsetSequentialSampler
np.random.seed(0)
indices = np.array([0, 2, 4, 6, 8, 10])

ARGS = {
    "indices": indices
}
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 SubsetSequentialSampler — 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 SubsetSequentialSampler 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
SubsetWeightedRandomSampler Ran
ran on a synthesized input · constructed an object; no result was computed
dsba-lab/openal/query_strategies/featmix_sampling.py
code served (permissive licence) · get_code("c3df65aa7a3cf216")
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, SubsetWeightedRandomSampler: 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 numpy as np
import torch
from torch.utils.data import Sampler

torch.manual_seed(0)

# Create small test data
indices = np.array([0, 1, 2, 3, 4])
weights = [0.1, 0.2, 0.3, 0.2, 0.2]
replacement = True

ARGS = {
    "indices": indices,
    "weights": weights,
    "replacement": replacement,
}
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 SubsetWeightedRandomSampler — 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 SubsetWeightedRandomSampler 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
TrainIterableDataset Ran
ran on a synthesized input · constructed an object; no result was computed
dsba-lab/openal/query_strategies/featmix_sampling.py
code served (permissive licence) · get_code("2e91b401d6cc9f05")
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, TrainIterableDataset: 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 numpy as np
import torch
from torch.utils.data import IterableDataset

torch.manual_seed(0)

# Create a minimal dataset stand-in with __len__ and __getitem__
class SimpleDataset:
    def __init__(self):
        self.data = [
            (torch.randn(4), torch.tensor(0)),
            (torch.randn(4), torch.tensor(1)),
            (torch.randn(4), torch.tensor(0)),
            (torch.randn(4), torch.tensor(1)),
            (torch.randn(4), torch.tensor(0)),
        ]
    
    def __len__(self):
        return len(self.data)
    
    def __getitem__(self, idx):
        return self.data[idx]

dataset = SimpleDataset()

ARGS = {
    "dataset": dataset,
    "sample_idx": np.array([0, 1, 2]),
    "return_index": True,
}
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 TrainIterableDataset — 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 TrainIterableDataset 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_target_from_dataset Ran
ran, honoured the contract we drafted
dsba-lab/openal/query_strategies/featmix_sampling.py
code served (permissive licence) · get_code("52d51edd564dc57a")
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 int, 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
import pandas as pd

# Create a mock ALDataset class
class ALDataset:
    def __init__(self):
        self.data_info = pd.DataFrame({
            'label': [0, 1, 0, 1]
        })

# Create a mock standard dataset class with 'targets' attribute
class StandardDataset:
    def __init__(self):
        self.targets = [0, 1, 0, 1]

# Create instances for testing
al_dataset = ALDataset()
standard_dataset = StandardDataset()

# Test with ALDataset
ARGS = {
    'dataset': al_dataset
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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 int, 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
AlphaMixSampling Not yet run
unverified (harvested, no recorded run)
aminparvaneh/alpha_mix_active_learning/query_strategies/alpha_mix_sampling.py
pointer only (licence: NONE) · get_code("6b8bd51846dfad76")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, aminparvaneh/alpha_mix_active_learning 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. No invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 5 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
Strategy
brick of this paper's assembly: helper
1 other paper
1 later
2210.07805 (2022-10)
Meta-Query-Net: Resolving Purity-Informativeness Dilemma in Open-set Active Learning
TrainIterableDataset
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2210.07805 (2022-10)
Meta-Query-Net: Resolving Purity-Informativeness Dilemma in Open-set Active Learning
get_target_from_dataset
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2210.07805 (2022-10)
Meta-Query-Net: Resolving Purity-Informativeness Dilemma in Open-set Active Learning

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 promise of active learning (AL) is to reduce labelling costs by selecting the most valuable examples to annotate from a pool of unlabelled data. Identifying these examples is especially challenging with high-dimensional data (e.g. images, videos) and in low-data regimes. In this paper, we propose a novel method for batch AL called ALFA-Mix. We identify unlabelled instances with sufficiently-distinct features by seeking inconsistencies in predictions resulting from interventions on their representations. We construct interpolations between representations of labelled and unlabelled instances then examine the predicted labels. We show that inconsistencies in these predictions help discovering features that the model is unable to recognise in the unlabelled instances. We derive an efficient implementation based on a closed-form solution to the optimal interpolation causing changes in predictions. Our method outperforms all recent AL approaches in 30 different settings on 12 benchmarks of images, videos, and non-visual data. The improvements are especially significant in low-data regimes and on self-trained vision transformers, where ALFA-Mix outperforms the stateof-the-art in 59% and 43% of the experiments respectively 1 .

For agents

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

get_harvested_code_for_paper("2203.07034")
get_code_for_paper("2203.07034")
have("2203.07034")

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

curl https://syntology.ai/api/ran/2203.07034.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.07034.svg)](https://syntology.ai/paper/2203.07034)

Connect an agent — have() is free.