Self-Routing Capsule Networks

Authors: Taeyoung Hahn, Myeongjang Pyeon, Gunhee Kim

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experiments on CIFAR10, SVHN, and Small NORB show that the self-routing performs more robustly against white-box adversarial attacks and affine transformations, requiring less computation.
Researcher Affiliation Academia Taeyoung Hahn Myeongjang Pyeon Gunhee Kim Seoul National University, Seoul, Korea
Pseudocode No No structured pseudocode or algorithm blocks were found.
Open Source Code Yes Our full source code is available at http://vision.snu.ac.kr/projects/self-routing.
Open Datasets Yes Datasets. Following Caps Net literature, we mostly use two classification benchmarks of CIFAR-10 [22] and SVHN [42] and additionally Small NORB [24] for the affine transformation tests.
Dataset Splits No The paper mentions 'training data' and 'test data' for specific experiments (e.g., '1/3 of training data' for Small NORB), but does not explicitly provide a general train/validation/test split for the datasets used.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory) used for running experiments were mentioned.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup No The paper states that 'more details of experiments... including architectures and optimization' are described in the Appendix, but these specific experimental setup details (e.g., hyperparameters, optimizer settings, batch size) are not present in the main text.