</> Code with Papers by Syntology

Papers › Instance-Dependent Noisy Label Learning via Graphical Modelling

Instance-Dependent Noisy Label Learning via Graphical Modelling

2 Sep 2022arXiv:2209.00906archive 2025-07-28

Arpit Garg, Cuong Nguyen, Rafael Felix, Thanh-Toan Do, Gustavo Carneiro

Noisy labels are unavoidable yet troublesome in the ecosystem of deep learning because models can easily overfit them. There are many types of label noise, such as symmetric, asymmetric and instance-dependent noise (IDN), with IDN being the only type that depends on image information. Such dependence on image information makes IDN a critical type of label noise to study, given that labelling mistakes are caused in large part by insufficient or ambiguous information about the visual classes present in images. Aiming to provide an effective technique to address IDN, we present a new graphical modelling approach called InstanceGM, that combines discriminative and generative models. The main contributions of InstanceGM are: i) the use of the continuous Bernoulli distribution to train the generative model, offering significant training advantages, and ii) the exploration of a state-of-the-art noisy-label discriminative classifier to generate clean labels from instance-dependent noisy-label samples. InstanceGM is competitive with current noisy-label learning approaches, particularly in IDN benchmarks using synthetic and real-world datasets, where our method shows better accuracy than the competitors in most experiments.

PaperPDFCode

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

Code

arpit2412/InstanceGM mentioned on GitHubpytorch 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.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image ClassificationLearning with noisy labels

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification Clothing1M InstanceGM Accuracy 74.40% #18 of 51 Archive leaderboard report
Image Classification Red MiniImageNet 20% label noise InstanceGM-SS Accuracy 60.89 #3 of 5 Archive leaderboard report
Image Classification Red MiniImageNet 20% label noise InstanceGM Accuracy 58.38 #4 of 5 Archive leaderboard report
Image Classification Red MiniImageNet 40% label noise InstanceGM-SS Accuracy 56.37 #2 of 5 Archive leaderboard report
Image Classification Red MiniImageNet 40% label noise InstanceGM Accuracy 52.24 #4 of 5 Archive leaderboard report
Image Classification Red MiniImageNet 60% label noise InstanceGM-SS Accuracy 53.21 #1 of 4 Archive leaderboard report
Image Classification Red MiniImageNet 60% label noise InstanceGM Accuracy 47.96 #3 of 4 Archive leaderboard report
Image Classification Red MiniImageNet 80% label noise InstanceGM-SS Accuracy 44.03 #2 of 5 Archive leaderboard report
Image Classification Red MiniImageNet 80% label noise InstanceGM Accuracy 39.62 #4 of 5 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM with ConvNeXt Accuracy 84.7 #11 of 19 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM with ConvNeXt ImageNet Pretrained NO #11 of 19 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM with ConvNeXt Network ConvNeXt #11 of 19 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM Accuracy 84.6 #12 of 19 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM ImageNet Pretrained NO #12 of 19 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM Network Vgg19-BN #12 of 19 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM with ResNet Accuracy 82.3 #15 of 19 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM with ResNet ImageNet Pretrained NO #15 of 19 Archive leaderboard report
Learning with noisy labels ANIMAL InstanceGM with ResNet Network ResNet #15 of 19 Archive leaderboard report
Learning with noisy labels CIFAR-10 InstanceGM Test Accuracy 95.9 #1 of 1 Archive leaderboard report
Learning with noisy labels CIFAR-100 InstanceGM Test Accuracy 77.19 #1 of 1 Archive leaderboard report
Learning with noisy labels Red MiniImageNet 20% label noise InstanceGM-SS Test Accuracy 60.89 #3 of 4 Archive leaderboard report
Learning with noisy labels Red MiniImageNet 20% label noise InstanceGM Test Accuracy 58.38 #4 of 4 Archive leaderboard report
Learning with noisy labels Red MiniImageNet 40% label noise InstanceGM-SS Test Accuracy 56.37 #3 of 4 Archive leaderboard report
Learning with noisy labels Red MiniImageNet 40% label noise InstanceGM Test Accuracy 52.24 #4 of 4 Archive leaderboard report
Learning with noisy labels Red MiniImageNet 60% label noise InstanceGM-SS Test Accuracy 53.21 #1 of 3 Archive leaderboard report
Learning with noisy labels Red MiniImageNet 60% label noise InstanceGM Test Accuracy 47.96 #3 of 3 Archive leaderboard report
Learning with noisy labels Red MiniImageNet 80% label noise InstanceGM-SS Test Accuracy 44.03 #2 of 4 Archive leaderboard report
Learning with noisy labels Red MiniImageNet 80% label noise InstanceGM Test Accuracy 39.62 #4 of 4 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