Interval Bound Interpolation for Few-shot Learning with Few Tasks

Authors: Shounak Datta, Sankha Subhra Mullick, Anish Chakrabarty, Swagatam Das

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

Reproducibility Variable Result LLM Response
Research Type Experimental The efficacy of our proposed approach is evident from the improved performance on several datasets from diverse domains compared to current methods. and In Section 5, we empirically demonstrate the effectiveness of our proposed approach, in comparison to the recent prior methods while making concluding remarks in Section 6.
Researcher Affiliation Academia 1 Electronics and Communication Sciences Unit, Indian Statistical Institute, Kolkata, India 2 Statistics and Mathematics Unit, Indian Statistical Institute, Kolkata, India.
Pseudocode Yes The steps for MAML+IBP/IBI and Proto Net+IBP/IBI are respectively presented in Algorithm 1 and 2.
Open Source Code Yes Further discussion on the datasets and implementation details of IBI along with the choice of hyperparameters can be found in the Appendix while the code is available at https://github.com/ Sankha Subhra/maml-ibp-ibi.
Open Datasets Yes The experiments are conducted on few-task few-shot image classification datasets, viz. a subset of the mini Image Net dataset called mini Image Net-S (Yao et al., 2022), and two medical images datasets namely Derm Net-S (Yao et al., 2022), and ISIC (Codella et al., 2018; Li et al., 2020). and Derm Net-S: Derm Net-S (Yao et al., 2022) is a subset of the Dermnet Skin Disease Atlas publicly available at http: //www.dermnet.com/.
Dataset Splits Yes Following the directives of Vinyals et al. (2016) from the total 100 classes, 64 are kept in the Training set, 16 are retained for validation, and the rest of the 20 classes are used for testing. and Moreover, we use random classes not included in the Training or Test set as the Validation set.
Hardware Specification Yes All the experiments are performed in the same environment using an RTX 3090 GPU.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies or libraries used in the experiments.
Experiment Setup Yes For hyperparameter tuning, we employ a grid search. In Table 10, we list the search spaces for each of the hyperparameters used in MAML+IBP and MAML+IBI. Moreover, in Table 11, we also detail the search spaces for each of the hyperparameters used in Proto Net+IBP and Proto Net+IBI.