Better Generalized Few-Shot Learning Even without Base Data

Authors: Seong-Woong Kim, Dong-Wan Choi

AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experimental results somewhat surprisingly show that the proposed zero-base GFSL method that does not utilize any base samples even outperforms the existing GFSL methods that make the best use of base data.
Researcher Affiliation Academia Seong-Woong Kim, Dong-Wan Choi* Department of Computer Science and Engineering, Inha University, South Korea wauri6@gmail.com, dchoi@inha.ac.kr
Pseudocode No The paper describes its methods in prose and mathematical formulas, but does not include any structured pseudocode or algorithm blocks.
Open Source Code Yes Our implementation is available at: https://github.com/bigdata-inha/Zero-Base-GFSL.
Open Datasets Yes We compare our method with the state-of-the-art (SOTA) GFSL methods using two datasets, mini-Image Net (Vinyals et al. 2016) and tiered-Image Net (Ren et al. 2018), which are most widely used in the literature of GFSL.
Dataset Splits Yes The mini-Image Net contains 100 classes and 60,000 sample images from Image Net (Russakovsky et al. 2015), which are then randomly split into 64 training classes, 16 validation classes, and 20 testing classes, proposed by (Ravi and Larochelle 2017).
Hardware Specification Yes We implement all the methods in Py Torch, and train each model on a machine with NVIDIA A100.
Software Dependencies No The paper mentions using "Py Torch" but does not provide specific version numbers for PyTorch or any other software dependencies.
Experiment Setup No The paper states "Full details of our settings are covered in Appendix." which implies that the specific experimental setup details like hyperparameters are not present in the main text.