Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..

Neural Collaborative Subspace Clustering

Authors: Tong Zhang, Pan Ji, Mehrtash Harandi, Wenbing Huang, Hongdong Li

ICML 2019 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We thoroughly assess and contrast the performance of our model against various state-of-the-art clustering algorithms including deep subspace-based ones.Our empirical study shows the superiority of the proposed algo-rithm over several state-of-the-art baselines including deep subspace clustering techniques.
Researcher Affiliation Collaboration 1Motovis Australia Pty Ltd 2Australian National University 3NEC Labs America 4Monash University 5Tencent AI Lab.
Pseudocode Yes Algorithm 1 Neural Collaborative Subspace Clustering
Open Source Code No The paper does not provide any explicit statement or link regarding the availability of open-source code for the described methodology.
Open Datasets Yes We evaluate our algorithm on three datasets , namely MNIST (Le Cun et al., 1998), Fashion-MNIST (Xiao et al., 2017), and the Stanford Online Products dataset (Oh Song et al., 2016)
Dataset Splits No The paper mentions training epochs but does not specify a separate validation dataset split with percentages or counts.
Hardware Specification Yes We implemented our framework with Tensorflow-1.6 (Abadi et al., 2016) on an Nvidia TITAN X GPU.
Software Dependencies Yes We implemented our framework with Tensorflow-1.6 (Abadi et al., 2016)
Experiment Setup Yes For all the experiments, we pre-train the convolutional auto-encoder for 60 epochs with a learning rate 1.0 10 3.We keep the λ1 = 10 in all the experiments, and slightly change the l and u for each dataset.We set the batch size to 5000, and used Adam (Kingma & Ba, 2014), an adaptive momentum based gradient descent method to minimize the loss in all our experiments.We set the learning rate to 1.0 10 5 for the auto-encoder and 1.0 10 3 for other parts of the network in all training stages.