Debiased Self-Training for Semi-Supervised Learning
Authors: Baixu Chen, Junguang Jiang, Ximei Wang, Pengfei Wan, Jianmin Wang, Mingsheng Long
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments justify that DST achieves an average improvement of 6.3% against state-of-the-art methods on standard semisupervised learning benchmark datasets and 18.9% against Fix Match on 13 diverse tasks. |
| Researcher Affiliation | Collaboration | Baixu Chen , Junguang Jiang , Ximei Wang, Pengfei Wan , Jianmin Wang, Mingsheng Long B School of Software, BNRist, Tsinghua University, China Y-tech, Kuaishou Technology |
| Pseudocode | No | The paper describes the proposed method using text and mathematical equations, and provides figures illustrating the process flow, but it does not include a formal pseudocode or algorithm block. |
| Open Source Code | Yes | We have released a benchmark containing both the code for our method and that for all the baselines at https://github.com/thuml/Debiased-Self-Training. |
| Open Datasets | Yes | Following [47], we construct a labeled subset with 4 labels per category to verify the effectiveness of DST in extremely label-scarce settings. To make a fair comparison, we keep the labeled subset for each dataset the same throughout our experiments. |
| Dataset Splits | No | The paper mentions evaluating on a 'class-balanced validation set' in Section 5.4 ('we calculate the class imbalance ratio I on a class-balanced validation set'), implying its use, but it does not provide specific details on the size, percentages, or method of splitting the dataset into training, validation, and test sets to enable reproduction of the data partitioning. |
| Hardware Specification | Yes | When training 1000k iterations on CIFAR-100 using 4 2080 Ti GPUs, Fix Match takes 104 hours while DST takes 111 hours, only a 7% increase in time. |
| Software Dependencies | No | The paper mentions using specific augmentation techniques like 'Rand Augment [11]' and optimizers like 'SGD with momentum 0.9', but it does not specify version numbers for any programming languages (e.g., Python), deep learning frameworks (e.g., PyTorch, TensorFlow), or other software libraries used in the experiments. |
| Experiment Setup | Yes | When training from scratch, we adopt the same hyperparameters as Fix Match [47], with learning rate of 0.03, mini-batch size of 512. For other experiments, we use SGD with momentum 0.9 and learning rates in {0.001, 0.003, 0.01, 0.03}. The mini-batch size is set to 64 following [49]. |