SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1608.07017 · 2016

Ambient Sound Provides Supervision for Visual Learning

arXiv · PDF · Open in the Atlas

Code that ran

11 of 13 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 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/1608.07017.json · badge: /api/ran/1608.07017.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
rowhanm/ambient-sound-self-supervision community (classifier label)
classifier basis not recorded on the link
11 ran · 2 no recorded run
FunctionStatusWhere it lives
alexnet Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/utils/models.py
code served (permissive licence) · get_code("170624d53a903496")
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 AlexNet. 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 math
import torch
import torch.nn as nn

torch.manual_seed(0)

CFG = {'2012': [(96, 11, 4, 2), 'M', (256, 5, 1, 2), 'M', (384, 3, 1, 1), (384, 3, 1, 1), (256, 3, 1, 1), 'M']}

class AlexNet(nn.Module):

    def __init__(self, features, num_classes, sobel):
        super(AlexNet, self).__init__()
        self.features = features
        self.classifier = nn.Sequential(nn.Dropout(0.5), nn.Linear(256 * 6 * 6, 4096), nn.ReLU(inplace=True), nn.Dropout(0.5), nn.Linear(4096, 4096), nn.ReLU(inplace=True))
        self.top_layer = nn.Linear(4096, num_classes)
        self._initialize_weights()
        if sobel:
            grayscale = nn.Conv2d(3, 1, kernel_size=1, stride=1, padding=0)
            grayscale.weight.data.fill_(1.0 / 3.0)
            grayscale.bias.data.zero_()
            sobel_filter = nn.Conv2d(1, 2, kernel_size=3, stride=1, padding=1)
            sobel_filter.weight.data[0, 0].copy_(torch.FloatTensor([[1, 0, -1], [2, 0, -2], [1, 0, -1]]))
            sobel_filter.weight.data[1, 0].copy_(torch.FloatTensor([[1, 2, 1], [0, 0, 0], [-1, -2, -1]]))
            sobel_filter.bias.data.zero_()
            self.sobel = nn.Sequential(grayscale, sobel_filter)
            for p in self.sobel.parameters():
                p.requires_grad = False
        else:
            self.sobel = None

    def forward(self, x):
        if self.sobel:
            x = self.sobel(x)
        x = self.features(x)
        x = x.view(x.size(0), 256 * 6 * 6)
        x = self.classifier(x)
        if self.top_layer:
            x = self.top_layer(x)
        return x

    def _initialize_weights(self):
        for y, m in enumerate(self.modules()):
            if isinstance(m, nn.Conv2d):
                n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
                for i in range(m.out_channels):
                    m.weight.data[i].normal_(0, math.sqrt(2.0 / n))
                if m.bias is not None:
                    m.bias.data.zero_()
            elif isinstance(m, nn.BatchNorm2d):
                m.weight.data.fill_(1)
                m.bias.data.zero_()
            elif isinstance(m, nn.Linear):
                m.weight.data.normal_(0, 0.01)
                m.bias.data.zero_()

def make_layers_features(cfg, input_dim, bn):
    layers = []
    in_channels = input_dim
    for v in cfg:
        if v == 'M':
            layers += [nn.MaxPool2d(kernel_size=3, stride=2)]
        else:
            conv2d = nn.Conv2d(in_channels, v[0], kernel_size=v[1], stride=v[2], padding=v[3])
            if bn:
                layers += [conv2d, nn.BatchNorm2d(v[0]), nn.ReLU(inplace=True)]
            else:
                layers += [conv2d, nn.ReLU(inplace=True)]
            in_channels = v[0]
    return nn.Sequential(*layers)

def alexnet(sobel=False, bn=True, out=1000):
    dim = 2 + int(not sobel)
    model = AlexNet(make_layers_features(CFG['2012'], dim, bn=bn), out, sobel)
    return model

ARGS = {
    'sobel': False,
    'bn': True,
    'out': 1000
}
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 AlexNet — a AlexNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a AlexNet 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
calculate_accuracy Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/pretext_training/pretext_train_5_alexnet.py
code served (permissive licence) · get_code("2789ed37f7a2637e")
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 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 torch
import torch.nn as nn

torch.manual_seed(0)

# fx: tensor of shape (batch_size, num_classes) containing logits/scores
# The function calls fx.max(1, keepdim=True)[1] to get predictions
# So fx needs to be a 2D tensor with batch dimension 0 and class dimension 1
fx = torch.randn(2, 4)

# y: tensor of shape (batch_size,) containing target class indices
# The function calls y.view_as(preds) where preds has shape (batch_size, 1)
# So y needs to be a 1D tensor that can be viewed as (batch_size, 1)
y = torch.tensor([0, 2])

ARGS = {
    "fx": fx,
    "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 Tensor, 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
deprocess_image Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/downstream_evaluation/grad_ascent.py
code served (permissive licence) · get_code("d7e0681d7b94b304")
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 3, dtype family int, shape (4, 5, 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
import torch

torch.manual_seed(0)

# Create a small 4D numpy array with shape (batch=1, channels=3, height=4, width=5)
# The function expects x[0, :, :, :] to work, so batch dimension must be present
x = np.random.randn(1, 3, 4, 5).astype('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, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (4, 5, 3) — shape (4, 5, 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
draw_on Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/utils/aolib_img.py
code served (permissive licence) · get_code("62a90a990e18bc83")
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 3, dtype family int, shape (100, 100, 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
from PIL import Image, ImageDraw

def to_pil(im):
    return Image.fromarray(np.uint8(im))

def from_pil(pil):
    return np.array(pil)

def draw_on(f, im):
    pil = to_pil(im)
    draw = ImageDraw.ImageDraw(pil)
    f(draw)
    return from_pil(pil)

# Create a simple drawing function that takes an ImageDraw object
def simple_draw_func(draw):
    draw.rectangle([10, 10, 50, 50], fill=(255, 0, 0))
    draw.line([0, 0, 100, 100], fill=(0, 255, 0), width=2)

# Create a simple RGB image (height=100, width=100, channels=3)
im = np.zeros((100, 100, 3), dtype=np.uint8)
im[:, :] = [200, 200, 200]  # light gray background

ARGS = {
    "f": simple_draw_func,
    "im": im
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (100, 100, 3) — shape (100, 100, 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
epoch_time Ran
ran, honoured the contract we drafted
rowhanm/ambient-sound-self-supervision/pretext_training/pretext_train_5_alexnet.py
pointer only (licence: MIT); the stored copy get_code serves, from takashiishida/flooding, is pointer only (licence: GPL-3.0) · get_code("3e21e6589663b136")
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 takashiishida/flooding, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

evaluate Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/pretext_training/pretext_train_5_alexnet.py
code served (permissive licence) · get_code("84d949f774e2ca7b")
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 1, dtype family float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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

torch.manual_seed(0)

# Build a small model that takes input and produces logits for classification
class SimpleModel(nn.Module):
    def __init__(self, input_dim, num_classes):
        super().__init__()
        self.linear = nn.Linear(input_dim, num_classes)
    
    def forward(self, x):
        return self.linear(x)

# Create model
model = SimpleModel(input_dim=4, num_classes=3)
model.eval()

# Create a simple data iterator (list of (x, y) tuples)
# x: batch of 2 samples, each with 4 features
# y: batch of 2 labels (class indices)
iterator = [
    (torch.randn(2, 4), torch.randint(0, 3, (2,))),
    (torch.randn(2, 4), torch.randint(0, 3, (2,))),
]

# Create criterion
criterion = nn.CrossEntropyLoss()

ARGS = {
    "model": model,
    "iterator": iterator,
    "criterion": criterion,
}
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 tuple, ndim 1, 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
freq2erb Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/pretext_training/get_statistical_summaries.py
code served (permissive licence) · get_code("c9706c092718039b")
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

def freq2erb(freq_hz):
    return 9.265 * np.log(1 + freq_hz / (24.7 * 9.265))

# freq_hz should be a numeric value or array-like that can be used with numpy operations
freq_hz_value = np.array([100.0, 440.0, 1000.0, 8000.0])

ARGS = {
    'freq_hz': freq_hz_value
}
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
make_layers_features Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/utils/models.py
code served (permissive licence) · get_code("68d8bd2d7e4d0e88")
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 Sequential. 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

torch.manual_seed(0)

# Construct cfg parameter - a list of layer specifications
# Each element is either 'M' for MaxPool or a tuple (out_channels, kernel_size, stride, padding)
cfg = [
    (16, 3, 1, 1),  # Conv2d: 3 input channels -> 16 output, kernel=3, stride=1, padding=1
    (32, 3, 1, 1),  # Conv2d: 16 -> 32
    'M',              # MaxPool2d
    (64, 3, 1, 1),  # Conv2d: 32 -> 64
    'M',              # MaxPool2d
]

input_dim = 3  # Input has 3 channels (e.g., RGB image)

bn = True  # Use batch normalization

ARGS = {
    'cfg': cfg,
    'input_dim': input_dim,
    'bn': bn,
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
rename_key Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/downstream_evaluation/finetune_voc.py
code served (permissive licence) · get_code("5f3cccf706a6e319")
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 str. 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)

def rename_key(key):
    if not 'module' in key:
        return key
    return ''.join(key.split('.module'))

# The function takes a single parameter 'key' which is a string
# It checks if 'module' is in the key string
# If not, it returns the key as-is
# If yes, it splits by '.module' and joins the parts back together

ARGS = {
    'key': 'model.module.layer.module.weight'
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
sound_from_subbands Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/pretext_training/get_statistical_summaries.py
code served (permissive licence) · get_code("aa0b8fa1809385c5")
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 (6,) 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

# Analyze the function:
# - subbands: used with np.fft.rfft(subbands, axis=0)
#   Must be a numpy array with at least 2 dimensions (since axis=1 is used in sum)
# - filts: multiplied with ft_subbands (element-wise)
#   Must have compatible shape for broadcasting with ft_subbands
# 
# subbands shape should be (time_steps, num_subbands)
# filts should be broadcastable to the shape of ft_subbands

# Create subbands: (6, 4) - 6 time steps, 4 subbands
subbands = np.random.randn(6, 4).astype(np.float32)

# Create filts: should have compatible shape for multiplication with rfft output
# rfft reduces the first dimension to n//2 + 1, so (4, 4) for (6, 4) input
# filts should be (4, 4) to broadcast correctly
filts = np.random.randn(4, 4).astype(np.float32)

ARGS = {
    "subbands": subbands,
    "filts": filts,
}
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 (6,) — shape (6,) 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
subbands_from_sound Ran
ran on a synthesized input
rowhanm/ambient-sound-self-supervision/pretext_training/get_statistical_summaries.py
code served (permissive licence) · get_code("716d4bc48005dd53")
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 (8, 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

# Construct arguments for subbands_from_sound

# signal: 1D numpy array representing audio signal
# The function applies rfft to it, so it needs to be a 1D array
signal = np.random.randn(8)

# filts: 2D numpy array of filters
# Based on the function body: ft_signal[:, np.newaxis] creates shape (freq_bins, 1)
# filts * ft_signal[:, np.newaxis] broadcasts filts with shape (freq_bins, num_filters)
# rfft of length 8 signal gives 5 frequency bins (8//2 + 1)
filts = np.random.randn(5, 3)

ARGS = {
    "signal": signal,
    "filts": filts,
}
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 (8, 3) — shape (8, 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
accuracy Not yet run
unverified (harvested, no recorded run)
rowhanm/ambient-sound-self-supervision/downstream_evaluation/finetune_voc.py
pointer only (licence: MIT); the stored copy get_code serves, from kkonstantinidis/ByzShield, is pointer only (licence: NONE) · get_code("4d3faaaa1c706b0b")
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 kkonstantinidis/ByzShield, and GitHub detected no licence in that repository 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.

load_model Not yet run
unverified (harvested, no recorded run)
rowhanm/ambient-sound-self-supervision/downstream_evaluation/ft_util.py
code served (permissive licence) · get_code("3abf2c16454d0592")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 3 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 3 later) and 10 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
accuracy 23 other papers
1 dated earlier, 1 the same month, 21 later
1603.07285 (2016-03)
A guide to convolution arithmetic for deep learning
1608.08710 (2016-08)
Pruning Filters for Efficient ConvNets
1803.09877 (2018-03)
DRACO: Byzantine-resilient Distributed Training via Redundant Gradients
and 20 more not named here
epoch_time 10 other papers
10 later
1911.05343 (2019-11)
A Stable Variational Autoencoder for Text Modelling
2002.08709 (2020-02)
Do We Need Zero Training Loss After Achieving Zero Training Error?
2010.12229 (2020-10)
Throughput-Optimal Topology Design for Cross-Silo Federated Learning
and 7 more not named here
make_layers_features 2 other papers
2 later
1911.05371 (2019-11)
Self-labelling via simultaneous clustering and representation learning
2103.13023 (2021-03)
Can Vision Transformers Learn without Natural Images?

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 sound of crashing waves, the roar of fast-moving cars -- sound conveys important information about the objects in our surroundings. In this work, we show that ambient sounds can be used as a supervisory signal for learning visual models. To demonstrate this, we train a convolutional neural network to predict a statistical summary of the sound associated with a video frame. We show that, through this process, the network learns a representation that conveys information about objects and scenes. We evaluate this representation on several recognition tasks, finding that its performance is comparable to that of other state-of-the-art unsupervised learning methods. Finally, we show through visualizations that the network learns units that are selective to objects that are often associated with characteristic sounds.

For agents

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

get_harvested_code_for_paper("1608.07017")
get_code_for_paper("1608.07017")
have("1608.07017")

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

curl https://syntology.ai/api/ran/1608.07017.json

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

[![Syntology run record](https://syntology.ai/api/ran/1608.07017.svg)](https://syntology.ai/paper/1608.07017)

Connect an agent — have() is free.