SC2Net: Sparse LSTMs for Sparse Coding
Authors: Joey Tianyi Zhou, Kai Di, Jiawei Du, Xi Peng, Hao Yang, Sinno Jialin Pan, Ivor Tsang, Yong Liu, Zheng Qin, Rick Siow Mong Goh
AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments show the effectiveness of our method on both unsupervised and supervised tasks. |
| Researcher Affiliation | Collaboration | Institute of High Performance Computing, A*STAR, Singapore, 2College of Computer Science, Sichuan University, China, 3Amazon, Seattle, USA, 4Nanyang Technological University, Singapore, 5University of Technology Sydney, Australia |
| Pseudocode | No | The paper describes the mathematical formulations and update rules for ISTA and SLSTM (e.g., equations 5, 6, 7, 8), but it does not present any explicit pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any explicit statements about releasing source code or links to a code repository for their proposed method. |
| Open Datasets | Yes | We evaluate the performance of these methods using MINST 2 and CIFAR-10 (Krizhevsky 2009). MINST contains 60,000 training images and 10,000 testing images sampled from the digits (0-9), where each image is of the size of 28 28. The CIFAR-10 dataset is another widely used benchmark, which contains 60,000 32 32 3 color images distributed over 10 subjects. |
| Dataset Splits | Yes | MINST contains 60,000 training images and 10,000 testing images sampled from the digits (0-9)... The CIFAR-10 dataset is another widely used benchmark, which contains 60,000 32 32 3 color images distributed over 10 subjects. |
| Hardware Specification | Yes | to train all neural-networkbased approaches with a GPU of NVIDIA TITAN X using Keras. |
| Software Dependencies | No | The paper mentions "using Keras" for training, but does not provide specific version numbers for Keras or any other software libraries or dependencies. |
| Experiment Setup | Yes | Thus, we fix the sparsity parameter λ = 0.1 for all evaluated methods in the unsupervised learning setting. |