Uncertainty-Aware Few-Shot Image Classification
Authors: Zhizheng Zhang, Cuiling Lan, Wenjun Zeng, Zhibo Chen, Shih-Fu Chang
IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments show our proposed method brings significant improvements on top of a strong baseline and achieves the state-of-the-art performance. |
| Researcher Affiliation | Collaboration | Zhizheng Zhang1 , Cuiling Lan2 , Wenjun Zeng2 , Zhibo Chen1 , Shih-Fu Chang3 1University of Science and Technology of China 2Microsoft Research Asia 3Columbia University |
| Pseudocode | No | The paper describes the method using textual explanations and mathematical equations, but does not include a dedicated pseudocode or algorithm block. |
| Open Source Code | No | The paper does not provide any explicit statement about releasing source code for the methodology or a link to a code repository. |
| Open Datasets | Yes | For few-shot image classification, we conduct experiments on four public benchmark datasets: mini Image Net [Vinyals et al., 2016], tiered-Image Net [Ren et al., 2018], CIFAR-FS [Bertinetto et al., 2018], and FC100 [Oreshkin et al., 2018]. |
| Dataset Splits | Yes | For few-shot image classification, all the categories/classes of the dataset are divided into base classes Cbase for training and novel classes Cnovel for testing without class overlapping [Snell et al., 2017; Chen et al., 2019; Dhillon et al., 2020]. In few-shot learning, episodic training is widely used. Each N-way K-shot task randomly sampled from base classes is defined as an episode, where the support set S includes N classes with K samples per class, and the query set Q contains the same N classes with M samples per class. We create each few-shot episode by uniformly sampling 5 classes (i.e.N=5) from the test set and further uniformly sampling support and query samples for each sampled class accordingly. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific software details with version numbers (e.g., Python 3.8, PyTorch 1.9, CUDA 11.1) needed to replicate the experiment. |
| Experiment Setup | Yes | For all our models, the classification temperature τ in (2) is a trainable parameter which is initialized with 10. For the parameter L, we found the performance is very similar when it is in the range of 16 to 64 and we set it to 32. We adopt the cross-entropy loss formulated by L = PN M i=1 yi log(p(yi|xi)), where p(yi|xi) is calculated in Eq. (2) with s( , ) instantiated by cosine similarity. |