Beyond Majority Voting: A Coarse-to-Fine Label Filtration for Heavily Noisy Labels

Han, Bo; Tsang, Ivor W.; Chen, Ling; Zhou, Joey Tianyi; Yu, Celina P.

doi:10.1109/tnnls.2019.2899045

Cited by 12 publications

(17 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ref. [51] proposed a coarse-to-fine label iteration model to dig out a set of high-quality labels from fully aggregated labels by using a sparse filter.…”

Section: Noisy Label Problemmentioning

confidence: 99%

Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China

et al. 2021

View full text Add to dashboard Cite

Large-scale land-cover classification using a supervised algorithm is a challenging task. Enormous efforts have been made to manually process and check the production of national land-cover maps. This has led to complex pre- and post-processing and even the production of inaccurate mapping products from large-scale remote sensing images. Inspired by the recent success of deep learning techniques, in this study we provided a feasible automatic solution for improving the quality of national land-cover maps. However, the application of deep learning to national land-cover mapping remains limited because only small-scale noisy labels are available. To this end, a mutual transfer network MTNet was developed. MTNet is capable of learning better feature representations by mutually transferring pre-trained models from time-series of data and fine-tuning current data. An interactive training strategy such as this can effectively alleviate the effects of inaccurate or noisy labels and unbalanced sample distributions, thus yielding a relatively stable classification system. Extensive experiments were conducted by focusing on several representative regions to evaluate the classification results of our proposed method. Quantitative results showed that the proposed MTNet outperformed its baseline model about 1%, and the accuracy can be improved up to 6.45% compared with the model trained by the training set of another year. We also visualized the national classification maps generated by MTNet for two different time periods to quantitatively analyze the performance gain. It was concluded that the proposed MTNet provides an efficient method for large-scale land cover mapping.

show abstract

“…Ref. [51] proposed a coarse-to-fine label iteration model to dig out a set of high-quality labels from fully aggregated labels by using a sparse filter.…”

Section: Noisy Label Problemmentioning

confidence: 99%

Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Although many research works have been proposed on these topics, they have some limitations yet and there also exist some unresolved issues. Specifically, previous works in learning with label noise claim that the "disagreement" strategy is crucial for alleviating the overfitting issue [27][28][29] [30]. Such framework suffers from computational inefficiency and does not hold learning consistency.…”

Section: Motivationsmentioning

confidence: 99%

“…Most of the successful deep learning algorithms are heavily dependent on ground-truth labels that are correctly assigned to the training dataset. However, it is usually challenging to collect such high-quality datasets with strong supervision information in many real-world scenarios [24,27].…”

Section: Thesis Organizationmentioning

confidence: 99%

“…Learning with noisy labels is a vital topic in weakly-supervised learning, attracting much recent interest with several directions. 1) Some methods propose to train on selected examples, using small-loss selection [27,59,60], GMM distribution [61,62] or (dis)agreement between two models [28,59,60]. 2) Some methods aim to design sample weighting schemes that give higher weights on clean examples [37,[63][64][65].…”

Section: M} With ỹJmentioning

confidence: 99%

“…However, it is challenging to estimate the noise transition matrix accurately. An alternative approach is training on selected or weighted samples, e.g., Mentornet [63], gradient-based reweight [65] and Co-teaching [27].…”

Section: Motivationmentioning

confidence: 99%

See 2 more Smart Citations

Natural robustness of machine learning in the open world

Wei¹

View full text Add to dashboard Cite

visor, Prof. Bo An. Without the great opportunity he provided, I could not have undertaken this Ph.D. journey, which has been a turning point in my life. During these years, he always maintains a gentle, patient, and kind style. He has not only provided a singularly suitable research environment with the greatest freedom, but also taught me the necessary knowledge in my research life, such as how to be professional and how to properly communicate with others. Words cannot express my gratitude to him. The experience in his group would definitely shape my future in many dimensions. I am so fortunate to become a Ph.D. student under his supervision and I wish to be such a nice professor like him in the future. I also had the pleasure of working with a group of talented people: Qingyu Guo,

show abstract