Adversarial Learning for Robust Deep Clustering

Authors: Xu Yang, Cheng Deng, Kun Wei, Junchi Yan, Wei Liu

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on two popular datasets show that the proposed adversarial learning method can significantly enhance the robustness and further improve the overall clustering performance.
Researcher Affiliation Collaboration 1School of Electronic Engineering, Xidian University, Xian 710071, China 2Department of CSE and Mo E Key Lab of Artificial Intelligence, Shanghai Jiao Tong University 3Tencent AI Lab, Shenzhen, China
Pseudocode Yes Algorithm 1 Adversarial Learning for Robust Deep Clustering
Open Source Code Yes The source code is available at https://github.com/xdxuyang/ALRDC.
Open Datasets Yes MNIST [18]: containing a total of 70,000 handwritten digits with 60,000 training and 10,000 testing samples, each being a 28 × 28 monochrome image. Fashion MNIST [31]: having the same number of images with the same image size as MNIST, but fairly more complicated.
Dataset Splits Yes MNIST [18]: containing a total of 70,000 handwritten digits with 60,000 training and 10,000 testing samples, each being a 28 × 28 monochrome image. Fashion MNIST [31]: having the same number of images with the same image size as MNIST, but fairly more complicated.
Hardware Specification No The paper does not provide specific details about the hardware used for experiments, such as GPU models, CPU types, or memory.
Software Dependencies No The paper does not provide specific version numbers for ancillary software dependencies (e.g., Python, PyTorch, TensorFlow, or other libraries).
Experiment Setup Yes In our experiments, we set λ = 1. The hyper-parameters β and γ are determined by different networks and datasets... For MNIST, the channel numbers and kernel sizes of the autoencoder network are the same as those in [37], and we employ one convolutional layer and three following residual blocks in the encoder for Fashion-MNIST. The clustering layers consist of four fully-connected layers, and Re LU is employed as nonlinear activation.