RegMixMatch: Optimizing Mixup Utilization in Semi-Supervised Learning

Authors: Haorong Han, Jidong Yuan, Chixuan Wei, Zhongyang Yu

AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results demonstrate that Reg Mix Match achieves state-of-the-art performance across various SSL benchmarks. ... In this section, we present an extensive experimental evaluation of the proposed Reg Mix Match method. We assess its performance across a variety of widely-used SSL datasets, including CIFAR-10/100 (Krizhevsky, Hinton et al. 2009), SVHN (Netzer et al. 2011), STL-10 (Coates, Ng, and Lee 2011), and Image Net (Deng et al. 2009), under different labeled data conditions. The experimental results are benchmarked against 13 established SSL algorithms... Additionally, we perform comprehensive ablation studies and hyperparameter analysis to validate the design choices behind Reg Mix Match.
Researcher Affiliation Academia 1Key Laboratory of Big Data and Artificial Intelligence in Transportation, Ministry of Education, China 2School of Computer Science and Technology, Beijing Jiaotong University 3School of Data Science and Intelligent Media, Communication University of China EMAIL, EMAIL
Pseudocode Yes Algorithm 1: Reg Mix Match Algorithm
Open Source Code Yes Code https://github.com/hhrd9/regmixmatch
Open Datasets Yes We assess its performance across a variety of widely-used SSL datasets, including CIFAR-10/100 (Krizhevsky, Hinton et al. 2009), SVHN (Netzer et al. 2011), STL-10 (Coates, Ng, and Lee 2011), and Image Net (Deng et al. 2009), under different labeled data conditions.
Dataset Splits Yes Dataset CIFAR10 CIFAR100 SVHN STL10 # Label 10 40 250 4000 400 2500 10000 40 1000 ... Specifically, Reg Mix Match achieves an error rate of only 4.35% on CIFAR-10 with 10 labels and 11.74% on STL-10 with 40 labels
Hardware Specification Yes The experiments are conducted on 2 24GB RTX 3090 GPU.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, CUDA versions). It only mentions model architectures like Wide Res Net-28-2 and Res Net-37-2.
Experiment Setup Yes For the implementation of Reg Mix Match, the threshold τc for the consistency loss is set in accordance with the Free Match approach, as adopted in the state-of-the-art Flat Match method. ... The values of τm, αh, and αl are set to 0.999, 1.0, and 16.0, respectively. ... As illustrated in Figure 5(a), the optimal value for τm is 0.999, which is higher than the threshold τc (0.95). ... the optimal values for αh and αl in Reg Mix Match are 1.0 and 16.0, respectively.