SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.10058 · ICCV · 2023

No Fear of Classifier Biases: Neural Collapse Inspired Federated Learning with Synthetic and Fixed Classifier

Tao Lin, Chao Wu, Zexi Li, Xinyi Shang, Rui He

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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 0 have no contract verdict; 2 failures 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/2303.10058.json · badge: /api/ran/2303.10058.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
zexilee/iccv-2023-fedetf marked official in Papers with Code
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
balanced_softmax_loss Ran
ran; our contract draft was wrong, not the code
zexilee/iccv-2023-fedetf/client_funct.py
code served (permissive licence) · get_code("a0ad76ccb9701aa3")
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)

# Small batch size and number of classes
batch_size = 2
num_classes = 4

# Create logits: [batch_size, num_classes]
logits = torch.randn(batch_size, num_classes)

# Create labels: [batch_size] with class indices
labels = torch.tensor([0, 2], dtype=torch.long)

# Create sample_per_class: [num_classes] with positive counts
sample_per_class = torch.tensor([10, 15, 20, 5], dtype=torch.long)

# Reduction mode
reduction = 'mean'

ARGS = {
    'logits': logits,
    'labels': labels,
    'sample_per_class': sample_per_class,
    'reduction': reduction
}
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
client_fedetf Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zexilee/iccv-2023-fedetf/client_funct.py
code served (permissive licence) · get_code("20e68ec6c3ab29f9")
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) and returned float, ndim 0, 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
import torch.nn.functional as F

torch.manual_seed(0)

class ProtoClassifier(nn.Module):
    def __init__(self, feature_dim, num_classes):
        super().__init__()
        self.proto = nn.Parameter(torch.randn(feature_dim, num_classes))

class SimpleModel(nn.Module):
    def __init__(self, feature_dim, num_classes):
        super().__init__()
        self.feature_dim = feature_dim
        self.num_classes = num_classes
        self.encoder = nn.Linear(4, feature_dim)
        self.proto_classifier = ProtoClassifier(feature_dim, num_classes)
        self.scaling_train = 1.0
    
    def forward(self, x):
        feature = self.encoder(x)
        return feature, None, None

class SimpleOptimizer:
    def __init__(self, model):
        self.model = model
    
    def zero_grad(self):
        for param in self.model.parameters():
            if param.grad is not None:
                param.grad.zero_()
    
    def step(self):
        pass

class SimpleNode:
    def __init__(self):
        self.model = SimpleModel(feature_dim=3, num_classes=5)
        self.optimizer = SimpleOptimizer(self.model)
        self.sample_per_class = torch.tensor([2, 3, 2, 1, 2], dtype=torch.long)
        
        data_batch_1 = (torch.randn(2, 4), torch.tensor([0, 1]))
        data_batch_2 = (torch.randn(2, 4), torch.tensor([2, 3]))
        self.local_data = [data_batch_1, data_batch_2]

class SimpleArgs:
    pass

args = SimpleArgs()
node = SimpleNode()

ARGS = {
    'args': args,
    'node': node,
    'opt': 'balancedloss',
    'loss': 0.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 — no output shape is stated: observe_signature.py could not observe it (call:AssertionError)
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 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 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 2 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 2026-11-22.

Repositories linked to this paper

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

Abstract

Data heterogeneity is an inherent challenge that hinders the performance of federated learning (FL). Recent studies have identified the biased classifiers of local models as the key bottleneck. Previous attempts have used classifier calibration after FL training, but this approach falls short in improving the poor feature representations caused by training-time classifier biases. Resolving the classifier bias dilemma in FL requires a full understanding of the mechanisms behind the classifier. Recent advances in neural collapse have shown that the classifiers and feature prototypes under perfect training scenarios collapse into an optimal structure called simplex equiangular tight frame (ETF). Building on this neural collapse insight, we propose a solution to the FL's classifier bias problem by utilizing a synthetic and fixed ETF classifier during training. The optimal classifier structure enables all clients to learn unified and optimal feature representations even under extremely heterogeneous data. We devise several effective modules to better adapt the ETF structure in FL, achieving both high generalization and personalization. Extensive experiments demonstrate that our method achieves state-ofthe-art performances on CIFAR-10, CIFAR-100, and Tiny-ImageNet. The code is available at https://github. com/ZexiLee/ICCV-2023-FedETF.

For agents

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

get_harvested_code_for_paper("2303.10058")
get_code_for_paper("2303.10058")
have("2303.10058")

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

curl https://syntology.ai/api/ran/2303.10058.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.10058.svg)](https://syntology.ai/paper/2303.10058)

Connect an agent — have() is free.