Fully Spiking Variational Autoencoder
Authors: Hiromichi Kamata, Yusuke Mukuta, Tatsuya Harada7059-7067
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We experimented with several datasets, and confirmed that it can generate images with the same or better quality compared to conventional ANNs. The code is available at https://github.com/kamata1729/Fully Spiking VAE. |
| Researcher Affiliation | Academia | Hiromichi Kamata 1, Yusuke Mukuta 1, Tatsuya Harada 1,2 1 The University of Tokyo 2 RIKEN {kamata, mukuta, harada}@mi.t.u-tokyo.ac.jp |
| Pseudocode | No | The paper describes its methods using mathematical equations and figures, but does not include explicit pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code is available at https://github.com/kamata1729/Fully Spiking VAE. |
| Open Datasets | Yes | We conducted experiments using MNIST (Deng 2012), Fashion MNIST (Xiao, Rasul, and Vollgraf 2017), CIFAR10 (Krizhevsky and Hinton 2009), and Celeb A (Liu et al. 2015) |
| Dataset Splits | No | For MNIST and Fashion MNIST, we used 60,000 images for training and 10,000 images for evaluation. The input images were resized to 32 × 32. For CIFAR10, we used 50,000 images for training and 10,000 images for evaluation. For Celeb A, we used 162,770 images for training and 19,962 images for evaluation. |
| Hardware Specification | No | The paper mentions neuromorphic devices like Loihi and True North as target platforms for SNNs, but does not specify the hardware (e.g., GPU/CPU models) used to run the experiments described in the paper. |
| Software Dependencies | No | We implemented FSVAE in PyTorch (Paszke et al. 2019) |
| Experiment Setup | Yes | We use AdamW optimizer (Loshchilov and Hutter 2019), which trains 150 epochs with a learning rate of 0.001 and a weight decay of 0.001. The batch size is 250. In prior model, teacher forcing (Williams and Zipser 1989) is used to stabilize training, so that the prior’s input is zq,t, which is sampled from the posterior model. In addition, to prevent posterior collapse, scheduled sampling (Bengio et al. 2015) is performed. With a certain probability, we input zp,t to the prior instead of zq,t. This probability varies linearly from 0 to 0.3 during training. |