Collaborative Learning by Detecting Collaboration Partners

Authors: Shu Ding, Wei Wang

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on real-world datasets verify the effectiveness of our method.
Researcher Affiliation Academia Shu Ding, Wei Wang National Key Laboratory for Novel Software Technology Nanjing University, Nanjing 210023, China {dings, wangw}@lamda.nju.edu.cn
Pseudocode Yes Algorithm 1 Adaptive Collaborative Learning with Modularity Maximization (ACLMM); Algorithm 2 Adaptive Collaborative Learning with Clustering (ACLC)
Open Source Code No The paper does not provide an explicit statement about releasing source code or a link to a code repository.
Open Datasets Yes The MNIST dataset consists of 70K handwritten digit images in 10 classes; The CIFAR-10/100 dataset consists of 60K color images in 10/100 classes; The Federated Extended MNIST (FEMNIST) dataset is built by partitioning the data in Extended MNIST based on the writer of the digit/character.
Dataset Splits Yes The MNIST dataset consists of 70K handwritten digit images in 10 classes, which has a training set of 60K examples and a test set of 10K examples. They have 50K training images and 10K test images.
Hardware Specification No The paper does not explicitly describe the hardware used for running its experiments, such as specific GPU or CPU models.
Software Dependencies No The paper mentions using CNN models and ResNet-18, but does not provide specific software dependency versions (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup Yes We set N = 10 and M = 2000 for MNIST. We set N = 20 and M = 5000 for CIFAR-10 and set N = 50 and M = 20000 for CIFAR-100. We further set the number of groups K = 3 for MNIST, K = 4 for CIFAR-10, K = 7 for CIFAR-100 and K = 4 for FEMNIST.