基于引领森林的多粒度广义长尾分类

doi:10.11896/jsjkx.231100112

Abstract

Abstract: Long-tailed classification is an inevitable and challenging task in the real world.Traditional methods usually focus only on inter-class imbalanced distributions,however,recent studies have begun to emphasize intra-class long-tailed distributions,i.e.,within the same class,there are far more samples with head attributes than tail ones.Due to the implicitness of the attributes and the complexity of their combinations,the intra-class imbalance problem is even more difficult to deal with.For this purpose,a generalized long-tailed classification framework(Cognisance) is proposed in the paper,aiming to build a multi-granularity joint solution model for the long-tailed classification problem through the invariant feature learning.Firstly,the framework constructs coarse-grained leading forest(CLF) through unsupervised learning to better characterize the distribution of samples about diffe-rent attributes within the class,and thus constructs different environments in the process of invariant risk minimization.Secondly,the framework designs a new metric learning loss,multi-center loss(MCL),to gradually eliminate confusing attributes during the feature learning process.Additionally,the framework does not depend on a specific model structure and can be integrated with other long-tailed classification methods as an independent component.Experimental results on datasets ImageNet-GLT and MSCOCO-GLT show that,the proposed method achieves the best performance,and existing methods all gain an improvement of 2%~8% in Top1-Accuracy metric by integrating with this framework.

Key words: Long-tailed classification, Imbalance learning, Invariant feature learning, Multi-granularity joint problem solving

CLC Number:

TP391

YANG Jinye, XU Ji, WANG Guoyin. Multi-granular and Generalized Long-tailed Classification Based on Leading Forest[J].Computer Science, 2024, 51(11): 229-238.

References

[1] ZHANG Y,KANG B,HOOI B,et al.Deep Long-Tailed Lear-ning:A Survey[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(9):10795-10816.
[2] YANG Y,WANG H,KATABI D.On multi-domain long-tailed recognition,imbalanced domain generalization and beyond[C]//Proceedings of European Conference on Computer Vision.Tel-Aviv,Israel,2022:57-75.
[3] BUDA M,MAKI A,MAZUROWSKI M A.A systematic study of the class imbalance problem in convolutional neural networks[J].Neural Networks,2018,106:249-259.
[4] ANDO S,HUANG C Y.Deep over-sampling framework forclassifying imbalanced data[C]//Machine Learning and Know-ledge Discovery in Databases:European Conference.Skopje,Macedonia,2017:18-22.
[5] YANG Y,ZHANG G,KATABI D,et al.Me-net:Towards ef-fective adversarial robustness with matrix estimation[C]//International Conference on Machine Learning.California,USA,2019:7025-7034.
[6] CAO K,WEI C,GAIDON A,et al.Learning imbalanced data-sets with label-distribution-aware margin loss[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems.Vancouver,Canada,2019:1567-1578.
[7] SUN L,WANG K,YANG K,et al.See clearer at night:towards robust nighttime semantic segmentation through day-nightimage conversion[C]//Artificial Intelligence and Machine Learning in Defense Applications.Strasbourg,France,2019:77-89.
[8] TANG K,TAO M,QI J,et al.Invariant feature learning forgeneralized long-tailed classification[C]//European Conference on Computer Vision.Tel-Aviv,Israel,2022:709-726.
[9] DRUMMOND C,HOLTE R C.C4.5,class imbalance,and cost sensitivity:why under-sampling beats over-sampling[C]//Proceedings of the ICML’03 Workshop on Learning from Imba-lanced Data Sets II.Washington,USA,2003:1-8.
[10] HAN H,WANG W Y,MAO B H.Borderline-SMOTE:a newover-sampling method in imbalanced data sets learning[C]//International Conference on Intelligent Computing.Berlin,Heidelberg,2005:878-887.
[11] HE H,GARCIA E A.Learning from imbalanced data[J].IEEETransactions on Knowledge and Data Engineering,2009,21(9):1263-1284.
[12] LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice,Italy,2017:2980-2988
[13] KANG B,XIE S,ROHRBACH M,et al.Decoupling Representation and Classifier for Long-Tailed Recognition[C]//International Conference on Learning Representations.Addis Ababa,Ethiopia,2020.
[14] ZHOU B,CUI Q,WEI X S,et al.Bbn:Bilateral-branch network with cumulative learning for long-tailed visual recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington,USA,2020:9719-9728.
[15] ZHU B,NIU Y,HUA X S,et al.Cross-domain empirical riskminimization for unbiased long-tailed classification[C]//Proceedings of the AAAI Conference on Artificial Intelligence.Vancouver,Canada,2022:3589-3597.
[16] CUI P,ATHEY S.Stable learning establishes some commonground between causal inference and machine learning[J].Nature Machine Intelligence,2022,4(2):110-115.
[17] ARJOVSKY M,BOTTOU L,GULRAJANI I,et al.Invariant risk minimization[J].arXiv:1907.02893,2019.
[18] CREAGER E,JACOBSEN J H,ZEMEL R.Environment infe-rence for invariant learning[C]//International Conference on Machine Learning.2021:2189-2200.
[19] WEN Y,ZHANG K,LI Z,et al.A discriminative feature lear-ning approach for deep face recognition[C]//Computer Vision－ECCV 2016:14th European Conference.Amsterdam,The Ne-therlands,2016:499-515.
[20] WANG T,LI Y,KANG B,et al.The devil is in classification:A simple framework for long-tail instance segmentation[C]//Computer Vision－ECCV 2020:16th European Conference.Glasgow,UK,2020:728-744.
[21] ESTABROOKS A,JO T,JAPKOWICZ N.A multiple resam-pling method for learning from imbalanced data sets[J].Computational Intelligence,2004,20(1):18-36.
[22] ZHANG Z,PFISTER T.Learning fast sample re-weightingwithout reward data[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.Montreal,Canada,2021:725-734.
[23] REN J,YU C,MA X,et al.Balanced meta-softmax for long-tailed visual recognition[J].Advances in Neural Information Processing Systems,2020,33:4175-4186.
[24] ELKAN C.The foundations of cost-sensitive learning[C]//International Joint Conference on Artificial Intelligence.Seattle,USA,2001:973-978.
[25] JAMAL M A,BROWN M,YANG M H,et al.Rethinking class-balanced methods for long-tailed visual recognition from a domain adaptation perspective[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle,USA,2020:7610-7619.
[26] TAN J,WNAG C,LI B,et al.Equalization loss for long-tailed object recognition[C]//Proceedings of the IEEE/CVF Confe-rence on Computer Vision and Pattern Recognition.Seattle,USA,2020:11662-11671.
[27] TAN J,LU X,ZHANG G,et al.Equalization loss v2:A new gradient balance approach for long-tailed object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Nashville,USA,2021:1685-1694.
[28] MENON A K,JAYASUMANA S,RAWAT A S,et al.Long-tail learning via logit adjustment[C]//International Conference on Learning Representations.Addis Ababa,Ethiopia,2020.
[29] CHU P,BIAN X,LIU S,et al.Feature space augmentation for long-tailed data[C]//Computer Vision－ECCV 2020:16th European Conference.Glasgow,UK,2020:694-710.
[30] WANG J,LUKASIEWICZ T,HU X,et al.Rsg:A simple but effective module for learning imbalanced datasets[C]//Procee-dings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Nashville,USA,2021:3784-3793.
[31] HU X,JIANG Y,TANG K,et al.Learning to segment the tail[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle,USA,2020:14045-14054.
[32] CUBUK E D,ZOPH B,SHLENS J,et al.Randaugment:Practical automated data augmentation with a reduced search space[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle,USA,2020:702-703.
[33] LIU J,SUN Y,HAN C,et al.Deep representation learning on long-tailed data:A learnable embedding augmentation perspective[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle,USA,2020:2970-2979.
[34] ZHANG H,CISSE M,DAUPHIN Y N,et al.mixup:BeyondEmpirical Risk Minimization[C]//International Conference on Learning Representations.Vancouver,Canada,2018.
[35] WANG X,LIAN L,MIAO Z,et al.Long-tailed Recognition by Routing Diverse Distribution-Aware Experts[C]//International Conference on Learning Representations.Addis Ababa,Ethiopia,2020.
[36] ZHANG Y,HOOI B,HONG L,et al.Self-supervised aggrega-tion of diverse experts for test-agnostic long-tailed recognition[J].Advances in Neural Information Processing Systems,2022,35:34077-34090.
[37] XU J,WANG G,DENG W.DenPEHC:Density peak based efficient hierarchical clustering[J].Information Sciences,2016,373:200-218.
[38] RODRIGUEZ A,LAIO A.Clustering by fast search and find of density peaks[J].Science,2014,344(6191):1492-1496.
[39] RUSSAKOVSKY O,DENG J,SU H,et al.Imagenet large scale visual recognition challenge[J].International Journal of Computer Vision,2015,115:211-252.
[40] PATTERSON G,HAYS J.Coco attributes:Attributes for people,animals,and objects[C]//Computer Vision－ECCV 2016:14th European Conference.Amsterdam,The Netherlands,2016:85-100.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Comments

Recommended 0

No Suggested Reading articles found!

Multi-granular and Generalized Long-tailed Classification Based on Leading Forest

PDF (PC)

Abstract

Cite this article

share this article

References

Related Articles 1

Metrics

Comments

Recommended 0