SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.04156 · NeurIPS · 2021

Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Loss

Adrien Gaidon, Colin Wei, Tengyu Ma, Jeff Haochen

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 1 failure of our instrument; 0 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/2106.04156.json · badge: /api/ran/2106.04156.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
jhaochenz/spectral_contrastive_learning community
Papers with Code archive
2 ran · 0 no recorded run
jhaochenz96/spectral_contrastive_learning community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
D Ran
ran; our contract draft was wrong, not the code
jhaochenz96/spectral_contrastive_learning/models/spectral.py
code served (permissive licence) · get_code("cd6d57e7c8432808")
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 numpy as np

torch.manual_seed(0)

# Create small tensor inputs
# z1 and z2 should be 2D tensors (batch_size, feature_dim)
# Using batch_size=2, feature_dim=4
z1 = torch.randn(2, 4)
z2 = torch.randn(2, 4)

# mu is a float parameter controlling the radius
mu = 1.0

ARGS = {
    'z1': z1,
    'z2': z2,
    'mu': mu
}
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 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
Spectral Ran
ran on a synthesized input · constructed an object; no result was computed
jhaochenz/spectral_contrastive_learning/models/spectral.py
code served (permissive licence) · get_code("86d68d2245cab2f2")
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, Spectral: 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 numpy as np
import torch.nn as nn
from torchvision.models import resnet50

torch.manual_seed(0)

# Create a minimal backbone that outputs small tensors
class MinimalBackbone(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(3, 4)
    
    def forward(self, x):
        # Expects input of shape (batch, 3)
        return self.linear(x)

# Create minimal projector
class MinimalProjector(nn.Module):
    def __init__(self):
        super().__init__()
    
    def set_layers(self, num_layers):
        pass
    
    def forward(self, x):
        return x

# Build the backbone
backbone = MinimalBackbone()

# Create input tensors: batch size 2, feature dim 3
x1 = torch.randn(2, 3)
x2 = torch.randn(2, 3)

ARGS = {
    'backbone': backbone,
    'mu': 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Spectral — 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 Spectral 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
projection_identity Ran
ran on a synthesized input
jhaochenz/spectral_contrastive_learning/models/spectral.py
code served (permissive licence) · get_code("55c30097ea15926c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 2, 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

class projection_identity(nn.Module):

    def __init__(self):
        super().__init__()

    def set_layers(self, num_layers):
        pass

    def forward(self, x):
        return x


INIT_ARGS = {}

ARGS = {
    'x': torch.randn(2, 4)
}

if __name__ == '__main__':
    model = projection_identity(**INIT_ARGS)
    output = model(**ARGS)
    print(output)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 2, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 later) and 1 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
projection_identity
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2308.05017 (2023-08)
When and How Does Known Class Help Discover Unknown Ones? Provable Understanding Through Spectral Analysis
2311.03524 (2023-11)
A Graph-Theoretic Framework for Understanding Open-World Semi-Supervised Learning
D
brick of this paper's assembly: entry, loss
1 other paper
1 later
2405.18635 (2024-05)
When and How Does In-Distribution Label Help Out-of-Distribution Detection?

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

Recent works in self-supervised learning have advanced the state-of-the-art by relying on the contrastive learning paradigm, which learns representations by pushing positive pairs, or similar examples from the same class, closer together while keeping negative pairs far apart. Despite the empirical successes, theoretical foundations are limited -prior analyses assume conditional independence of the positive pairs given the same class label, but recent empirical applications use heavily correlated positive pairs (i.e., data augmentations of the same image). Our work analyzes contrastive learning without assuming conditional independence of positive pairs using a novel concept of the augmentation graph on data. Edges in this graph connect augmentations of the same datapoint, and ground-truth classes naturally form connected sub-graphs. We propose a loss that performs spectral decomposition on the population augmentation graph and can be succinctly written as a contrastive learning objective on neural net representations. Minimizing this objective leads to features with provable accuracy guarantees under linear probe evaluation. By standard generalization bounds, these accuracy guarantees also hold when minimizing the training contrastive loss. Empirically, the features learned by our objective can match or outperform several strong baselines on benchmark vision datasets. In all, this work provides the first provable analysis for contrastive learning where guarantees for linear probe evaluation can apply to realistic empirical settings.

For agents

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

get_harvested_code_for_paper("2106.04156")
get_code_for_paper("2106.04156")
have("2106.04156")

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

curl https://syntology.ai/api/ran/2106.04156.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.04156.svg)](https://syntology.ai/paper/2106.04156)

Connect an agent — have() is free.