August 24, 2024Open Access

RC-Mixup: A Data Augmentation Strategy against Noisy Data for Regression Tasks

View Full Paper

Ask AI

Bookmark

Authors

SHSeong-Hyeon HwangKorea Advanced Institute of Science and Technology MKMinsu KimKorea Advanced Institute of Science and Technology SWSteven Euijong WhangKorea Advanced Institute of Science and Technology

Discussion

Member takes

Implication

Computational evaluation demonstrates that RC-Mixup improves regression performance on noisy datasets, suggesting data-centric augmentation enhances multi-round robust model training.

Key Points

RC-Mixup outperforms C-Mixup and baseline robust training methods across noisy data benchmarks by selectively filtering and interpolating training samples for regression tasks.
The framework tightly integrates distance-based C-Mixup with multi-round robust training, allowing clean sample identification to enhance data augmentation without modifying algorithms.
Benchmark evaluation confirms that combining robust training with C-Mixup creates a synergistic data-centric pipeline, significantly improving model generalization under noisy conditions.

Cite This Study

Hwang et al. (2024) studied this question.

synapsesocial.com/papers/68e5b13eb6db64358754aae1 https://doi.org/10.1145/3637528.3671993

View Full Paper

Ask AI

Bookmark

Implication

Computational evaluation demonstrates that RC-Mixup improves regression performance on noisy datasets, suggesting data-centric augmentation enhances multi-round robust model training.

RC-Mixup: A Data Augmentation Strategy against Noisy Data for Regression Tasks

Discussion

Member takes

Implication

Key Points

Cite This Study

RC-Mixup: A Data Augmentation Strategy against Noisy Data for Regression Tasks

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider