Stacked Capsule Autoencoders
Authors: Adam Kosiorek, Sara Sabour, Yee Whye Teh, Geoffrey E. Hinton
NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We find that object capsule presences are highly informative of the object class, which leads to state-of-the-art results for unsupervised classification on SVHN (55%) and MNIST (98.7%). |
| Researcher Affiliation | Collaboration | Adam R. Kosiorek adamk@robots.ox.ac.uk Sara Sabour Yee Whye Teh Geoffrey E. Hinton Applied AI Lab Oxford Robotics Institute University of Oxford Department of Statistics University of Oxford Google Brain Toronto Deep Mind London This work was done during an internship at Google Brain. |
| Pseudocode | No | The paper describes the model components and their interactions using prose and mathematical equations, but it does not include any explicit pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code is available at github.com/google-research/google-research/tree/master/stacked_capsule_autoencoders. |
| Open Datasets | Yes | We train SCAE on MNIST, SVHN8 and CIFAR10 and try to assign class labels to vectors of object capsule presences. |
| Dataset Splits | Yes | In agreement with previous work on unsupervised clustering (Ji et al., 2018; Hu et al., 2017; Hjelm et al., 2019; Haeusser et al., 2018), we train our models and report results on full datasets (TRAIN, VALID and TEST splits). |
| Hardware Specification | No | The paper does not specify the hardware used for its experiments, such as particular GPU or CPU models, or memory configurations. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions) that would be needed to replicate the experiment. |
| Experiment Setup | Yes | We used an PCAE with 24 single-channel 11 11 templates for MNIST and 24 and 32 three-channel 14 14 templates for SVHN and CIFAR10, respectively. The OCAE used 24, 32 and 64 object capsules, respectively. Further details on model architectures and hyper-parameter tuning are available in Appendix A. |