Papers › Class-Balanced Loss Based on Effective Number of Samples

Class-Balanced Loss Based on Effective Number of Samples

16 Jan 2019CVPR 2019 6arXiv:1901.05555archive 2025-07-28

Yin Cui, Menglin Jia, Tsung-Yi Lin, Yang song, Serge Belongie

With the rapid increase of large-scale, real-world datasets, it becomes critical to address the problem of long-tailed data distribution (i.e., a few classes account for most of the data, while most classes are under-represented). Existing solutions typically adopt class re-balancing strategies such as re-sampling and re-weighting based on the number of observations for each class. In this work, we argue that as the number of samples increases, the additional benefit of a newly added data point will diminish. We introduce a novel theoretical framework to measure data overlap by associating with each sample a small neighboring region rather than a single point. The effective number of samples is defined as the volume of samples and can be calculated by a simple formula (1-βⁿ)/(1-β), where n is the number of samples and β∈[0,1) is a hyperparameter. We design a re-weighting scheme that uses the effective number of samples for each class to re-balance the loss, thereby yielding a class-balanced loss. Comprehensive experiments are conducted on artificially induced long-tailed CIFAR datasets and large-scale datasets including ImageNet and iNaturalist. Our results show that when trained with the proposed class-balanced loss, the network is able to achieve significant performance gains on long-tailed datasets.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 27 code samples harvested from 6 repositories linked to this paper: 15 ran (of which 2 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 2 where Syntology's instrument failed) · 12 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 19 samples from 1 repository, 8 ran (8 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 8 samples from 5 repositories, 7 ran (5 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 4 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

richardaecn/class-balanced-loss officialmentioned in papermentioned on GitHubtfMITSyntology: 19 samples, 8 ran (0 instrument-failed) · 11 unverified report
bazinga699/ncl mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (1 instrument-failed) · 0 unverified report
feidfoe/AdjustBnd4Imbalance mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Adjusting Decision Boundary for Class Imbalanced Learning (arXiv:1912.01857; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
frgfm/Holocron mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
lijian16/fcc mentioned on GitHubpytorchMITSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
statsu1990/yoto_class_balanced_loss mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
tiagoCuervo/JapaNet mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Deep Residual Learning for Image Recognition (arXiv:1512.03385; that paper's own run record: 254 ran (of which 108 constructed an object rather than computing a result; 163 with no instrument failure: 3 honoured, 1 violated, 159 with no contract checked; 91 where Syntology's instrument failed) · 123 unverified) and Searching for MobileNetV3 (arXiv:1905.02244; that paper's own run record: 86 ran (of which 22 constructed an object rather than computing a result; 75 with no instrument failure: 9 honoured, 2 violated, 64 with no contract checked; 11 where Syntology's instrument failed) · 19 unverified); nothing checks that they implement this paper's method. report
vandit15/Class-balanced-loss-pytorch mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (1 instrument-failed) · 1 unverified report
MindCode-4/code-11 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Entropy Law: The Story Behind Data Compression and LLM Performance (arXiv:2407.06645; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
MindCode-4/code-6 mindsporeSyntology: not harvested report
MindSpore-scientific/code-3 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Coneheads: Hierarchy Aware Attention (arXiv:2306.00392; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 6 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

27 samples harvested: 15 ran (of which 2 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 2 where Syntology's instrument failed) · 12 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · our draft was wrong
13ran
12unverified

Licence: 0 of the 27 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 6 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 1 sample that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

richardaecn/class-balanced-loss official repository 19 samples, 8 ran (0 instrument-failed) · 11 unverified
convert_array_to_image tpu/models/experimental/dcgan/cifar_input.py ran MIT (permissive) · 44a9936773ea082c · report fixture: signature array:ndarray/3/float · returned Image · level recorded 2026-09-25
dir2version src/cifar_main.py ran MIT (permissive) · 90a8002ad4ce5c44 · report fixture: signature data_dir:str/0/str · returned str · level recorded 2026-09-25
get_imbalanced_data src/generate_cifar_tfrecords_im.py ran MIT (permissive) · ea12a62c609bc291 · report fixture: signature training_data:dict/0/?, img_num_per_cls:dict/0/? · returned dict · level recorded 2026-09-25
img_num src/data_utils.py ran MIT (permissive) · 4f59b56fcbea8270 · report fixture: signature cifar_version:str/0/str · returned int · level recorded 2026-09-25
learning_rate_schedule in 15 papers src/cifar_main.py ran MIT (permissive) · 10fc348fea580bf2 · report fixture: signature current_epoch:EagerTensor/0/float, base_learning_rate:float/0/float, lr_boundaries:list/1/int, lr_multiplier:list/1/float · returned EagerTensor · level recorded 2026-09-25
local_device_setter src/cifar_utils.py ran MIT (permissive) · 667cea6bb6608b7b · report fixture: signature num_devices:int/0/int, ps_device_type:str/0/str, worker_device:str/0/str, ps_ops:list/1/str, ps_strategy:callable/0/? · returned function · level recorded 2026-09-25
read_json in 250 papers src/data_utils.py ran MIT (permissive) · 35ae8c0dce858108 · report fixture: signature data_dir:str/0/str · returned dict · level recorded 2026-09-25
sort_input src/generate_cifar_tfrecords_im.py ran MIT (permissive) · c20afab5ec1e1d08 · report fixture: signature im_data:dict/0/?, cifar_version:str/0/str · returned dict · level recorded 2026-09-25
discriminator in 10 papers tpu/models/experimental/dcgan/cifar_model.py unverified MIT (permissive) · 59ec4e09ac2ce079 · report
generate_input_fn tpu/models/experimental/dcgan/dcgan_main.py unverified MIT (permissive) · 787ada018be76773 · report
generator in 11 papers tpu/models/experimental/dcgan/cifar_model.py unverified MIT (permissive) · 6bfcb692805e9db9 · report
get_traindata_list src/generate_cifar_tfrecords_im.py unverified MIT (permissive) · 306ffaa84f8da0c6 · report
input_fn in 7 papers tpu/models/experimental/cifar_keras/cifar_keras.py unverified MIT (permissive) · 57184f395fe4ff8f · report
model_fn in 10 papers tpu/models/experimental/cifar_keras/cifar_keras.py unverified MIT (permissive) · a8d0a5e19105636a · report
model_fn in 10 papers tpu/models/experimental/dcgan/dcgan_main.py unverified MIT (permissive) · 7634c2c31837e3f5 · report
noise_input_fn tpu/models/experimental/dcgan/dcgan_main.py unverified MIT (permissive) · f3d3946d06874d2a · report
parser in 5 papers tpu/models/experimental/dcgan/cifar_input.py unverified MIT (permissive) · 1863a653dad9ca35 · report
read_pickle_from_file src/data_utils.py unverified MIT (permissive) · 162b6ba3306530cc · report
read_pickle_from_file src/generate_cifar_tfrecords.py unverified MIT (permissive) · 7137e8ae47a966ce · report
bazinga699/ncl community (archive-listed) 2 samples, 2 ran (1 instrument-failed) · 0 unverified
NBOD lib/loss/loss.py ran · our draft was wrong BSD-2-Clause (permissive) · e4d180e212ea7092 · report fixture: signature inputs:list/1/Tensor, factor:float/0/float · returned Tensor · level recorded 2026-09-22
NIL_NBOD lib/loss/loss.py ran BSD-2-Clause (permissive) · 49b09b16b775ec58 · report fixture: signature inputs:list/1/Tensor, targets:list/1/Tensor · returned Tensor · level recorded 2026-09-24
frgfm/Holocron community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
ClassBalancedWrapper holocron/nn/modules/loss.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · fa48de5142876479 · report fixture: signature x:Tensor/2/float, target:Tensor/1/int · returned Tensor · level recorded 2026-09-22
lijian16/fcc community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
ClassBalanceCE lib/loss/loss_impl/class_balanced_loss.py ran fingerprinted MIT (permissive) · 749ae093d1002dc8 · report fixture: returned ClassBalanceCE · level recorded 2026-09-22
CrossEntropy lib/loss/loss_impl/class_balanced_loss.py ran MIT (permissive) · 03ec3d933bc9e286 · report fixture: signature inputs:Tensor/2/float, targets:Tensor/1/int · returned Tensor · level recorded 2026-09-24
statsu1990/yoto_class_balanced_loss community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
ClassBalanced_CELoss src/model/cb_loss.py ran MIT (permissive) · 68d1f62eac95f880 · report fixture: returned ClassBalanced_CELoss · level recorded 2026-09-22
vandit15/Class-balanced-loss-pytorch community (archive-listed) 2 samples, 1 ran (1 instrument-failed) · 1 unverified
CB_loss class_balanced_loss.py ran · our draft was wrong MIT (permissive) · 213dab5ee3c9a730 · report fixture: signature labels:Tensor/1/int, logits:Tensor/2/float, samples_per_cls:list/1/int, no_of_classes:int/0/int, loss_type:str/0/str, beta:float/0/float, gamma:float/0/float · returned Tensor · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

5 assemblies on this paper: 5 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

5 sibling assemblies for Class-Balanced Loss Based on Effective Number of Samples (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Class-Balanced Loss Based on Effective Number of Samples (graph method: Class-Balanced Loss Based on Effective Number of Samples). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 5 siblings

  1. ClassBalanced_CELoss entry · level 1 · ran · 68d1f62eac95f880

harvested Assembly for Class-Balanced Loss Based on Effective Number of Samples (graph method: Class-Balanced Loss Based on Effective Number of Samples). 2 bricks; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 5 siblings

  1. CB_loss entry · level 1 · ran · our draft was wrong · 213dab5ee3c9a730
  2. focal_loss loss · level 0 · unverified · c32ccdb4e101c470

harvested Assembly for Class-Balanced Loss Based on Effective Number of Samples (graph method: Class-Balanced Loss Based on Effective Number of Samples). 2 bricks; individually on a fixture: 2 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 5 siblings

  1. ClassBalanceCE entry · level 1 · ran · 749ae093d1002dc8
  2. CrossEntropy helper · level 1 · ran · 03ec3d933bc9e286

harvested Assembly for Class-Balanced Loss Based on Effective Number of Samples (graph method: Class-Balanced Loss Based on Effective Number of Samples). 2 bricks; individually on a fixture: 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 5 siblings

  1. NIL_NBOD entry · level 1 · ran · 49b09b16b775ec58
  2. NBOD helper · level 1 · ran · our draft was wrong · e4d180e212ea7092

harvested Assembly for Class-Balanced Loss Based on Effective Number of Samples (graph method: Class-Balanced Loss Based on Effective Number of Samples). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 5 siblings

  1. ClassBalancedWrapper entry · level 1 · ran · fa48de5142876479

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="1901.05555") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image ClassificationLong-tail Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification iNaturalist 2018 ResNet-152 Top-1 Accuracy 69.05% #41 of 60 Archive leaderboard report
Image Classification iNaturalist 2018 ResNet-101 Top-1 Accuracy 67.98% #44 of 60 Archive leaderboard report
Image Classification iNaturalist 2018 ResNet-50 Top-1 Accuracy 64.16% #52 of 60 Archive leaderboard report
Long-tail Learning CIFAR-10-LT (ρ=10) Class-balanced Focal Loss Error Rate 12.90 #45 of 50 Archive leaderboard report
Long-tail Learning CIFAR-10-LT (ρ=10) Class-balanced Reweighting Error Rate 13.46 #48 of 50 Archive leaderboard report
Long-tail Learning CIFAR-100-LT (ρ=100) Cross-Entropy (CE) Error Rate 61.68 #65 of 66 Archive leaderboard report
Long-tail Learning COCO-MLT CB Loss(ResNet-50) Average mAP 49.06 #9 of 13 Archive leaderboard report
Long-tail Learning EGTEA CB Loss Average Precision 63.39 #2 of 3 Archive leaderboard report
Long-tail Learning EGTEA CB Loss Average Recall 63.26 #2 of 3 Archive leaderboard report
Long-tail Learning VOC-MLT CB Focal(ResNet-50) Average mAP 75.24 #9 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections