SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.16189 · NeurIPS · 2020

Self-supervised learning through the eyes of a child

Brenden Lake, A Orhan, Vaibhav Gupta

arXiv · PDF · Open in the Atlas

Code that ran

7 of 11 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; 1 failure of our instrument; 4 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/2007.16189.json · badge: /api/ran/2007.16189.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
eminorhan/baby-vision canonical (classifier label)
classifier basis: classifier_2025
0 ran · 1 no recorded run
agentic-learning-ai-lab/memory-storyboard community (classifier label)
classifier basis not recorded on the link
7 ran · 2 no recorded run
copy not recorded community
not in the paper's declared repositories
0 ran · 1 no recorded run
FunctionStatusWhere it lives
build_norm_layer Ran
ran on a synthesized input
agentic-learning-ai-lab/memory-storyboard/resnet.py
code served (permissive licence) · get_code("7d7aaaa31a292818")
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. 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)

# Test case 1: BatchNorm2d with default settings
ARGS = {
    'cfg': {
        'type': 'BN',
        'eps': 1e-05,
    },
    'num_features': 64,
    'postfix': '0',
}
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
count_seen_classes Ran
ran on a synthesized input
agentic-learning-ai-lab/memory-storyboard/dataset_utils.py
code served (permissive licence) · get_code("3313ad502a167458")
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 int; 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

torch.manual_seed(0)

ARGS = {
    "video_dirs": ["class1,class2", "class3,class4,class5", "class6"]
}
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 list, ndim 1, dtype family int — 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
get_svm_perf Ran
ran on a synthesized input
agentic-learning-ai-lab/memory-storyboard/evaluate.py
code served (permissive licence) · get_code("15218baa9671626b")
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 float, 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 numpy as np
from sklearn.linear_model import SGDClassifier

# Create small synthetic datasets
np.random.seed(0)

# train_features: shape (10, 4) - 10 samples, 4 features
train_features = np.random.randn(10, 4).astype(np.float32)

# test_features: shape (5, 4) - 5 samples, 4 features
test_features = np.random.randn(5, 4).astype(np.float32)

# train_labels: binary labels for 10 samples
train_labels = np.array([0, 1, 0, 1, 0, 1, 0, 1, 0, 1], dtype=np.int32)

# test_labels: binary labels for 5 samples
test_labels = np.array([0, 1, 0, 1, 0], dtype=np.int32)

ARGS = {
    "train_features": train_features,
    "test_features": test_features,
    "train_labels": train_labels,
    "test_labels": test_labels,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
load_split_train_test Ran
ran on a synthesized input
agentic-learning-ai-lab/memory-storyboard/linear_decoding.py
code served (permissive licence) · get_code("1f53b17b0b8cce8d")
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. 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 os
import tempfile
import torch
import torch.utils.data
import torchvision.transforms as transforms
from PIL import Image
import numpy as np

# Create a minimal args object
class Args:
    def __init__(self):
        self.image_size = 224
        self.batch_size = 2
        self.workers = 0

# Create temporary directory structure with dummy image data
temp_dir = tempfile.mkdtemp()
train_dir = os.path.join(temp_dir, 'train')
val_dir = os.path.join(temp_dir, 'val')

# Create class subdirectories
os.makedirs(os.path.join(train_dir, 'class0'), exist_ok=True)
os.makedirs(os.path.join(train_dir, 'class1'), exist_ok=True)
os.makedirs(os.path.join(val_dir, 'class0'), exist_ok=True)
os.makedirs(os.path.join(val_dir, 'class1'), exist_ok=True)

# Create dummy images for training
for i in range(3):
    img = Image.fromarray(np.random.randint(0, 256, (256, 256, 3), dtype=np.uint8))
    img.save(os.path.join(train_dir, 'class0', f'img_{i}.jpg'))
    img = Image.fromarray(np.random.randint(0, 256, (256, 256, 3), dtype=np.uint8))
    img.save(os.path.join(train_dir, 'class1', f'img_{i}.jpg'))

# Create dummy images for validation
for i in range(2):
    img = Image.fromarray(np.random.randint(0, 256, (256, 256, 3), dtype=np.uint8))
    img.save(os.path.join(val_dir, 'class0', f'img_{i}.jpg'))
    img = Image.fromarray(np.random.randint(0, 256, (256, 256, 3), dtype=np.uint8))
    img.save(os.path.join(val_dir, 'class1', f'img_{i}.jpg'))

ARGS = {
    'datadir': temp_dir,
    'args': Args()
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — 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
make_res_layer Ran
ran on a synthesized input
agentic-learning-ai-lab/memory-storyboard/resnet.py
code served (permissive licence) · get_code("6e21bb915de7277a")
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)

# Define a minimal block class that satisfies the interface expected by make_res_layer
class MinimalBlock(nn.Module):
    expansion = 2  # Used in make_res_layer to compute output channels
    
    def __init__(self, inplanes, planes, stride=1, dilation=1, downsample=None, 
                 activation='relu', norm_cfg=None, **kwargs):
        super().__init__()
        self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=stride, 
                               padding=dilation, dilation=dilation, bias=False)
        self.norm_cfg = norm_cfg
        self.downsample = downsample
        self.stride = stride
        self.activation = activation
    
    def forward(self, x):
        return x

# Build the norm_cfg dict
norm_cfg_dict = dict(type='BN')

# Define ARGS for make_res_layer
ARGS = {
    'block': MinimalBlock,
    'inplanes': 4,
    'planes': 8,
    'blocks': 2,
    'stride': 1,
    'dilation': 1,
    'norm_cfg': norm_cfg_dict,
    'activation': 'relu'
}
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
off_diagonal Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
agentic-learning-ai-lab/memory-storyboard/osiris_model.py
pointer only (licence: MIT); the stored copy get_code serves, from neurai-lab/tamil, is pointer only (licence: NONE) · get_code("3e30d88eaef01190")
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 neurai-lab/tamil, 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. A run is recorded.

train Ran
ran on a synthesized input
agentic-learning-ai-lab/memory-storyboard/linear_decoding.py
code served (permissive licence) · get_code("9e8c24c5b951741f")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. 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, 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
import torch.optim as optim

torch.manual_seed(0)

# Define a minimal model that has eval_forward method
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def eval_forward(self, x):
        return self.linear(x)
    
    def forward(self, x):
        return self.eval_forward(x)

# Define a minimal args class with print_freq
class Args:
    def __init__(self):
        self.print_freq = 2

# Create a minimal train loader with 2 batches
train_loader = [
    (torch.randn(2, 4), torch.randint(0, 3, (2,))),
    (torch.randn(2, 4), torch.randint(0, 3, (2,)))
]

# Create model, criterion, and optimizer
model = MinimalModel()
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.01)

# Create args
args = Args()

ARGS = {
    'train_loader': train_loader,
    'model': model,
    'criterion': criterion,
    'optimizer': optimizer,
    'epoch': 1,
    'args': args
}
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 float, 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
accuracy Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from zbh2047/clipping-algorithms
pointer only; the stored copy get_code serves, from zbh2047/clipping-algorithms, is pointer only (licence: NONE) · get_code("9b8289076669fe4f")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

get_all_fns Not yet run
unverified (harvested, no recorded run)
agentic-learning-ai-lab/memory-storyboard/dataset_utils.py
code served (permissive licence) · get_code("0a7e98cab5aea2a7")
How it ran

no invocation recorded.

get_knn_perf Not yet run
unverified (harvested, no recorded run)
agentic-learning-ai-lab/memory-storyboard/evaluate.py
code served (permissive licence) · get_code("1c2b9fdf428c7961")
How it ran

no invocation recorded.

train Not yet run
unverified (harvested, no recorded run)
eminorhan/baby-vision/temporal_classification.py
code served (permissive licence) · get_code("68e213dd1bd7c775")
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 (2 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
accuracy 80 other papers
42 dated earlier, 38 later
1307.0414 (2013-07)
Challenges in Representation Learning: A report on three machine learning contests
1412.6553 (2014-12)
Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition
1503.02531 (2015-03)
Distilling the Knowledge in a Neural Network
and 77 more not named here
off_diagonal 23 other papers
2 dated earlier, 21 later · 5 only as an assembly brick
1908.10084 (2019-08)
Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
2004.04136 (2020-04)
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
2103.03230 (2021-03)
Barlow Twins: Self-Supervised Learning via Redundancy Reduction
and 20 more not named here

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

Within months of birth, children develop meaningful expectations about the world around them. How much of this early knowledge can be explained through generic learning mechanisms applied to sensory data, and how much of it requires more substantive innate inductive biases? Addressing this fundamental question in its full generality is currently infeasible, but we can hope to make real progress in more narrowly defined domains, such as the development of high-level visual categories, thanks to improvements in data collecting technology and recent progress in deep learning. In this paper, our goal is precisely to achieve such progress by utilizing modern self-supervised deep learning methods and a recent longitudinal, egocentric video dataset recorded from the perspective of three young children (Sullivan et al., 2020). Our results demonstrate the emergence of powerful, high-level visual representations from developmentally realistic natural videos using generic selfsupervised learning objectives.

For agents

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

get_harvested_code_for_paper("2007.16189")
get_code_for_paper("2007.16189")
have("2007.16189")

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

curl https://syntology.ai/api/ran/2007.16189.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.16189.svg)](https://syntology.ai/paper/2007.16189)

Connect an agent — have() is free.