SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1707.04585 · 2017

The Reversible Residual Network: Backpropagation Without Storing Activations

arXiv · PDF · Open in the Atlas

Code that ran

4 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; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 0 no recorded run), measured 2026-11-21; 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/1707.04585.json · badge: /api/ran/1707.04585.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
renmengye/revnet-public canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
gigantenbein/ReVAE extension (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
evaluate Ran
ran, honoured the contract we drafted
renmengye/revnet-public/run_cifar_train.py
code served (permissive licence) · get_code("90df0fb80b7e2878")
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 float64, 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
import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a minimal model that implements infer_step method
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)  # 4 input features, 3 classes
    
    def infer_step(self, sess, img):
        """
        Takes sess and img batch, returns predictions of shape (batch_size, num_classes)
        """
        # img is expected to be a numpy array or tensor of shape (batch_size, 4)
        if isinstance(img, np.ndarray):
            img = torch.from_numpy(img).float()
        logits = self.linear(img)
        return logits.detach().numpy()

# Create a minimal session object (not used in the function, but required as parameter)
class SimpleSession:
    pass

# Create data iterator with batches containing 'img' and 'label' keys
data_iter = [
    {
        'img': np.array([[0.1, 0.2, 0.3, 0.4], [0.5, 0.6, 0.7, 0.8]], dtype=np.float32),
        'label': np.array([0, 1], dtype=np.int64)
    },
    {
        'img': np.array([[0.2, 0.3, 0.4, 0.5], [0.6, 0.7, 0.8, 0.9]], dtype=np.float32),
        'label': np.array([1, 2], dtype=np.int64)
    }
]

ARGS = {
    'sess': SimpleSession(),
    'model': SimpleModel(),
    'data_iter': data_iter
}
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 float64, 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
loss_function Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
gigantenbein/ReVAE/model.py
code served (permissive licence) · get_code("87522c59fd90aded")
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.functional as F

torch.manual_seed(0)

# The function expects:
# - recon_x: reconstructed output from VAE, shape (batch_size, 3072) after view
# - x: original input, will be viewed to (-1, 3072)
# - mu: mean from encoder, used with .size() and .pow(2)
# - logvar: log variance from encoder, used with .size() and .exp()

# For a batch size of 2, and since 3072 is the flattened size (32x32x3 images),
# we need recon_x and x to be compatible with that reshape.
# We can work with shape (2, 3072) directly or (2, 3, 32, 32) which reshapes to (2, 3072).

batch_size = 2
latent_dim = 4

# Create recon_x: reconstructed data, shape (batch_size, 3072)
recon_x = torch.sigmoid(torch.randn(batch_size, 3072))

# Create x: original data, shape (batch_size, 3072) or (batch_size, 3, 32, 32)
# Using (batch_size, 3072) for simplicity
x = torch.rand(batch_size, 3072)

# Create mu: mean from encoder, shape (batch_size, latent_dim)
mu = torch.randn(batch_size, latent_dim)

# Create logvar: log variance from encoder, shape (batch_size, latent_dim)
logvar = torch.randn(batch_size, latent_dim)

ARGS = {
    "recon_x": recon_x,
    "x": x,
    "mu": mu,
    "logvar": logvar,
}
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
train_step Ran
ran, honoured the contract we drafted
renmengye/revnet-public/run_cifar_train.py
code served (permissive licence) · get_code("afcb9f0b0da3af03")
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 torch
import torch.nn as nn

torch.manual_seed(0)

class MockSession:
    """Mock TensorFlow-like session object."""
    pass

class SimpleModel(nn.Module):
    """Minimal model that implements train_step method."""
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def train_step(self, sess, img, label):
        """Train step that accepts session, image, and label tensors."""
        output = self.linear(img)
        loss = nn.functional.mse_loss(output, label)
        return loss.item()

# Create instances
sess = MockSession()
model = SimpleModel()

# Create batch with distinct tensor shapes
batch = {
    'img': torch.randn(2, 4),      # batch_size=2, feature_dim=4
    'label': torch.randn(2, 3)      # batch_size=2, num_classes=3
}

ARGS = {
    'sess': sess,
    'model': model,
    'batch': batch
}
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
train_step Ran
ran, honoured the contract we drafted
renmengye/revnet-public/run_imagenet_train.py
code served (permissive licence) · get_code("c5abb3c9365dc91c")
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)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def train_step(self, sess):
        # This method is called by the train_step function
        # Return a cross-entropy loss value
        return torch.tensor(0.5)

class SimpleSession:
    def __init__(self):
        self.state = {}

sess = SimpleSession()
model = SimpleModel()

ARGS = {
    "sess": sess,
    "model": model
}
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

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, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 4 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.

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Deep residual networks (ResNets) have significantly pushed forward the state-of-the-art on image classification, increasing in performance as networks grow both deeper and wider. However, memory consumption becomes a bottleneck, as one needs to store the activations in order to calculate gradients using backpropagation. We present the Reversible Residual Network (RevNet), a variant of ResNets where each layer's activations can be reconstructed exactly from the next layer's. Therefore, the activations for most layers need not be stored in memory during backpropagation. We demonstrate the effectiveness of RevNets on CIFAR-10, CIFAR-100, and ImageNet, establishing nearly identical classification accuracy to equally-sized ResNets, even though the activation storage requirements are independent of depth.

For agents

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

get_harvested_code_for_paper("1707.04585")
get_code_for_paper("1707.04585")
have("1707.04585")

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

curl https://syntology.ai/api/ran/1707.04585.json

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

[![Syntology run record](https://syntology.ai/api/ran/1707.04585.svg)](https://syntology.ai/paper/1707.04585)

Connect an agent — have() is free.