CircConv: A Structured Convolution with Low Complexity
Authors: Siyu Liao, Bo Yuan4287-4294
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Through extensive experiments, such strong structureimposing approach is proved to be able to substantially reduce the number of parameters of convolutional layers and enable significant saving of computational cost by using fast multiplication of the circulant tensor. |
| Researcher Affiliation | Academia | Siyu Liao, Bo Yuan Department of Electrical and Computer Engineering, Rutgers University siyu.liao@rutgers.edu, bo.yuan@soe.rutgers.edu |
| Pseudocode | No | The paper does not contain any pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | No | The paper does not provide any statements or links regarding the public availability of its source code. |
| Open Datasets | Yes | We evaluate our circulant structure-imposing approaches on two typical image classification datasets: CIFAR-10 (Krizhevsky and Hinton 2009) and Image Net ILSVRC-2012 (Deng et al. 2009). |
| Dataset Splits | No | The paper mentions data augmentation for training and reports test error, but does not explicitly describe a validation split or its size/usage. |
| Hardware Specification | Yes | All models in this paper are trained using NVIDIA Ge Force GTX 1080 GPUs and Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz. |
| Software Dependencies | No | The paper mentions optimizers like SGD and RMSprop but does not specify any software libraries (e.g., TensorFlow, PyTorch) or their version numbers. |
| Experiment Setup | Yes | The compressed Res Net-32 models are trained using stochastic gradient descent (SGD) optimizer with learning rate 0.1, momentum 0.9, batch size 64 and weight decay 0.0001. |