Rethinking Self-Distillation: Label Averaging and Enhanced Soft Label Refinement with Partial Labels

• URL: http://arxiv.org/abs/2402.10482v2
• Date: Wed, 19 Feb 2025 10:04:02 GMT
• Title: Rethinking Self-Distillation: Label Averaging and Enhanced Soft Label Refinement with Partial Labels
• Authors: Hyeonsu Jeong, Hye Won Chung,
• Abstract summary: Multi-round self-distillation effectively performs label averaging among instances with high feature correlations. We introduce a novel, efficient single-round self-distillation method using refined partial labels from the teacher's top two softmax outputs.
• Score: 10.696635172502141
• License:
• Abstract: We investigate the mechanisms of self-distillation in multi-class classification, particularly in the context of linear probing with fixed feature extractors where traditional feature learning explanations do not apply. Our theoretical analysis reveals that multi-round self-distillation effectively performs label averaging among instances with high feature correlations, governed by the eigenvectors of the Gram matrix derived from input features. This process leads to clustered predictions and improved generalization, mitigating the impact of label noise by reducing the model's reliance on potentially corrupted labels. We establish conditions under which multi-round self-distillation achieves 100% population accuracy despite label noise. Furthermore, we introduce a novel, efficient single-round self-distillation method using refined partial labels from the teacher's top two softmax outputs, referred to as the PLL student model. This approach replicates the benefits of multi-round distillation in a single round, achieving comparable or superior performance--especially in high-noise scenarios--while significantly reducing computational cost.

Related papers

• Mitigating Instance-Dependent Label Noise: Integrating Self-Supervised Pretraining with Pseudo-Label Refinement [3.272177633069322]
  Real-world datasets often contain noisy labels due to human error, ambiguity, or resource constraints during the annotation process. We propose a novel framework that combines self-supervised learning using SimCLR with iterative pseudo-label refinement. Our approach significantly outperforms several state-of-the-art methods, particularly under high noise conditions.
  arXiv  Detail & Related papers  (2024-12-06T09:56:49Z)
• Dual-Decoupling Learning and Metric-Adaptive Thresholding for Semi-Supervised Multi-Label Learning [81.83013974171364]
  Semi-supervised multi-label learning (SSMLL) is a powerful framework for leveraging unlabeled data to reduce the expensive cost of collecting precise multi-label annotations. Unlike semi-supervised learning, one cannot select the most probable label as the pseudo-label in SSMLL due to multiple semantics contained in an instance. We propose a dual-perspective method to generate high-quality pseudo-labels.
  arXiv  Detail & Related papers  (2024-07-26T09:33:53Z)
• Extracting Clean and Balanced Subset for Noisy Long-tailed Classification [66.47809135771698]
  We develop a novel pseudo labeling method using class prototypes from the perspective of distribution matching. By setting a manually-specific probability measure, we can reduce the side-effects of noisy and long-tailed data simultaneously. Our method can extract this class-balanced subset with clean labels, which brings effective performance gains for long-tailed classification with label noise.
  arXiv  Detail & Related papers  (2024-04-10T07:34:37Z)
• SLaM: Student-Label Mixing for Distillation with Unlabeled Examples [15.825078347452024]
  We present a principled method for knowledge distillation with unlabeled examples that we call Student-Label Mixing (SLaM) SLaM consistently improves over prior approaches by evaluating it on several standard benchmarks. We give an algorithm improving the best-known sample complexity for learning halfspaces with margin under random classification noise.
  arXiv  Detail & Related papers  (2023-02-08T00:14:44Z)
• Exploiting Completeness and Uncertainty of Pseudo Labels for Weakly Supervised Video Anomaly Detection [149.23913018423022]
  Weakly supervised video anomaly detection aims to identify abnormal events in videos using only video-level labels. Two-stage self-training methods have achieved significant improvements by self-generating pseudo labels. We propose an enhancement framework by exploiting completeness and uncertainty properties for effective self-training.
  arXiv  Detail & Related papers  (2022-12-08T05:53:53Z)
• Leveraging Instance Features for Label Aggregation in Programmatic Weak Supervision [75.1860418333995]
  Programmatic Weak Supervision (PWS) has emerged as a widespread paradigm to synthesize training labels efficiently. The core component of PWS is the label model, which infers true labels by aggregating the outputs of multiple noisy supervision sources as labeling functions. Existing statistical label models typically rely only on the outputs of LF, ignoring the instance features when modeling the underlying generative process.
  arXiv  Detail & Related papers  (2022-10-06T07:28:53Z)
• Efficient and Flexible Sublabel-Accurate Energy Minimization [62.50191141358778]
  We address the problem of minimizing a class of energy functions consisting of data and smoothness terms. Existing continuous optimization methods can find sublabel-accurate solutions, but they are not efficient for large label spaces. We propose an efficient sublabel-accurate method that utilizes the best properties of both continuous and discrete models.
  arXiv  Detail & Related papers  (2022-06-20T06:58:55Z)
• Optimizing Diffusion Rate and Label Reliability in a Graph-Based Semi-supervised Classifier [2.4366811507669124]
  The Local and Global Consistency (LGC) algorithm is one of the most well-known graph-based semi-supervised (GSSL) classifiers. We discuss how removing the self-influence of a labeled instance may be beneficial, and how it relates to leave-one-out error. Within this framework, we propose methods to estimate label reliability and diffusion rate.
  arXiv  Detail & Related papers  (2022-01-10T16:58:52Z)
• Deep Semi-supervised Knowledge Distillation for Overlapping Cervical Cell Instance Segmentation [54.49894381464853]
  We propose to leverage both labeled and unlabeled data for instance segmentation with improved accuracy by knowledge distillation. We propose a novel Mask-guided Mean Teacher framework with Perturbation-sensitive Sample Mining. Experiments show that the proposed method improves the performance significantly compared with the supervised method learned from labeled data only.
  arXiv  Detail & Related papers  (2020-07-21T13:27:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.