Few-Shot Learning with Part Discovery and Augmentation from Unlabeled Images
Authors: Wentao Chen, Chenyang Si, Wei Wang, Liang Wang, Zilei Wang, Tieniu Tan
IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct systematic studies on mini Image Net and tiered Image Net benchmarks. Remarkably, our method yields impressive results, outperforming the previous best unsupervised methods by 7.74% and 9.24% under 5-way 1-shot and 5-way 5-shot settings, which are comparable with state-of-the-art supervised methods. |
| Researcher Affiliation | Academia | Wentao Chen1,2 , Chenyang Si2 , Wei Wang2 , Liang Wang2 , Zilei Wang1 , Tieniu Tan1,2 1University of Science and Technology of China 2Center for Research on Intelligent Perception and Computing, NLPR, CASIA {wentao.chen, chenyang.si}@cripac.ia.ac.cn, {wangwei, wangliang, tnt}@nlpr.ia.ac.cn, zlwang@ustc.edu.cn |
| Pseudocode | No | The paper does not contain any pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available. |
| Open Datasets | Yes | mini Image Net is a standard benchmark for few-shot learning proposed by [Vinyals et al., 2016]. It is a subset of the Image Net [Russakovsky et al., 2015] and contains 100 classes and 600 examples for each class. We follow the protocol in [Ravi and Larochelle, 2017] to use 64 classes for training, 16 classes for validation and 20 classes for test. tiered Image Net [Ren et al., 2018] is a larger subset of Image Net and contains 608 classes and 1000 images in each class. |
| Dataset Splits | Yes | We follow the protocol in [Ravi and Larochelle, 2017] to use 64 classes for training, 16 classes for validation and 20 classes for test. Theses classes are grouped into 34 higher categories, where 20 categories (351 classes) for training, 6 categories (97 classes) for validation and 8 categories (160 classes) for test. |
| Hardware Specification | No | The paper does not explicitly describe the hardware used to run its experiments (e.g., specific GPU/CPU models or types). |
| Software Dependencies | No | The paper mentions optimizers like SGD and Adam, but does not provide specific version numbers for any software components, libraries, or frameworks used (e.g., Python, PyTorch, TensorFlow versions). |
| Experiment Setup | Yes | For PDN, we transform input images with random crop, horizontal flip, color jitter and Gaussian blur. The crop scales for part view and global view are in range of (0.05, 0.14) and (0.14, 1), respectively. The number of cropped parts is set as n = 6 by cross validation. The size of D is 1024 and 10240 for mini Image Net and tiered Image Net, respectively. We use standard Res Net-50 as backbone and resize images to 224 224. We set hyper-parameter m = 0.999 and τ = 0.2. We adopt SGD optimizer with cosine learning rate decay. The learning rate is 0.015 for mini Image Net and 0.03 for tiered Image Net. For PAN, we retrieve Na = 1024 extra images for each class. The label smoothing hyper-parameter ϵ is 0.2 for 1shot and 0.7 for 5-shot. The loss weight λ is set as 1. The classifier is trained with Adam and the learning rate is 0.001. |