SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1905.11786 · 2019

Putting An End to End-to-End: Gradient-Isolated Learning of Representations

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 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 2 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/1905.11786.json · badge: /api/ran/1905.11786.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
loeweX/Greedy_InfoMax canonical (classifier label)
classifier basis: census
3 ran · 1 no recorded run
FunctionStatusWhere it lives
genOrthgonal Ran
ran, honoured the contract we drafted
loeweX/Greedy_InfoMax/GreedyInfoMax/utils/model_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from yanivbl6/BeyondSigProp, is pointer only (licence: NONE) · get_code("3f736c50902c4519")
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 yanivbl6/BeyondSigProp, 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.

reload_weights Ran
ran on a synthesized input
loeweX/Greedy_InfoMax/GreedyInfoMax/utils/model_utils.py
code served (permissive licence) · get_code("4788aa6d84fb2b52")
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
import tempfile
import os

torch.manual_seed(0)

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

# Define a simple model for audio experiment
class SimpleAudioModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Define a model with encoder for non-audio experiment
class SimpleEncoderModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.module = nn.Module()
        self.module.encoder = nn.ModuleList([
            nn.Linear(4, 3),
            nn.Linear(3, 2)
        ])
    
    def forward(self, x):
        for layer in self.module.encoder:
            x = layer(x)
        return x

# Create options object
class Options:
    def __init__(self, model_type, experiment, model_path, model_num, start_epoch, device):
        self.model_type = model_type
        self.experiment = experiment
        self.model_path = model_path
        self.model_num = model_num
        self.start_epoch = start_epoch
        self.device = device

# Create device
device = torch.device('cpu')

# Test case 1: Audio model with reload_model=True
opt_audio = Options(
    model_type=0,
    experiment='audio',
    model_path=temp_dir,
    model_num=1,
    start_epoch=0,
    device=device
)

model_audio = SimpleAudioModel()
optimizer_audio = torch.optim.Adam(model_audio.parameters())

# Save dummy checkpoint for audio
dummy_state = model_audio.state_dict()
torch.save(dummy_state, os.path.join(temp_dir, 'model_1.ckpt'))

ARGS = {
    'opt': opt_audio,
    'model': model_audio,
    'optimizer': optimizer_audio,
    'reload_model': True
}
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
val_by_InfoNCELoss Ran
ran on a synthesized input
loeweX/Greedy_InfoMax/GreedyInfoMax/audio/validation/val_by_InfoNCELoss.py
code served (permissive licence) · get_code("a039271bfdab388d")
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 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)

class SimpleModel(nn.Module):
    def __init__(self, model_splits):
        super().__init__()
        self.model_splits = model_splits
        self.linear = nn.Linear(4, model_splits)
    
    def forward(self, x):
        # x shape: (batch_size, 4)
        # output shape: (batch_size, model_splits)
        return self.linear(x)

class OptConfig:
    def __init__(self):
        self.device = 'cpu'
        self.model_splits = 3

# Create a minimal test loader with 2 batches
# Each batch is a tuple of (audio, _, _, _)
test_loader = [
    (torch.randn(2, 4), torch.tensor([0]), torch.tensor([1]), torch.tensor([2])),
    (torch.randn(2, 4), torch.tensor([0]), torch.tensor([1]), torch.tensor([2])),
]

opt = OptConfig()
model = SimpleModel(model_splits=opt.model_splits)

ARGS = {
    'opt': opt,
    'model': model,
    'test_loader': test_loader,
}
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 float — 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
distribute_over_GPUs Not yet run
unverified (harvested, no recorded run)
loeweX/Greedy_InfoMax/GreedyInfoMax/utils/model_utils.py
code served (permissive licence) · get_code("fcdd3f17c00b4c2a")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 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 3 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
genOrthgonal 1 other paper
1 later
2007.01038 (2020-07)
Beyond Signal Propagation: Is Feature Diversity Necessary in Deep Neural Network Initialization?

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

We propose a novel deep learning method for local self-supervised representation learning that does not require labels nor end-to-end backpropagation but exploits the natural order in data instead. Inspired by the observation that biological neural networks appear to learn without backpropagating a global error signal, we split a deep neural network into a stack of gradient-isolated modules. Each module is trained to maximally preserve the information of its inputs using the InfoNCE bound from Oord et al. [2018]. Despite this greedy training, we demonstrate that each module improves upon the output of its predecessor, and that the representations created by the top module yield highly competitive results on downstream classification tasks in the audio and visual domain. The proposal enables optimizing modules asynchronously, allowing large-scale distributed training of very deep neural networks on unlabelled datasets.

For agents

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

get_harvested_code_for_paper("1905.11786")
get_code_for_paper("1905.11786")
have("1905.11786")

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

curl https://syntology.ai/api/ran/1905.11786.json

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

[![Syntology run record](https://syntology.ai/api/ran/1905.11786.svg)](https://syntology.ai/paper/1905.11786)

Connect an agent — have() is free.