Scale Up Nonlinear Component Analysis with Doubly Stochastic Gradients

Authors: Bo Xie, Yingyu Liang, Le Song

NeurIPS 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate the effectiveness and scalability of our algorithm on large scale synthetic and real world datasets.
Researcher Affiliation Academia Bo Xie1, Yingyu Liang2, Le Song1 1Georgia Institute of Technology bo.xie@gatech.edu, lsong@cc.gatech.edu 2Princeton University yingyul@cs.princeton.edu
Pseudocode Yes Algorithm 1: {αi}t 1 = DSGD-KPCA(P(x), k)
Open Source Code No The paper does not contain any explicit statement about releasing the source code for the described methodology, nor does it provide a link to a code repository.
Open Datasets Yes Molecular Space dataset contains 2.3 million molecular motifs [6].
Dataset Splits Yes We randomly select 20% as test set and out of the remaining training set, we randomly choose 5000 as validation set to select step sizes.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments, such as GPU/CPU models or memory.
Software Dependencies No The paper does not list specific software dependencies with version numbers.
Experiment Setup Yes In each iteration, we use a data mini-batch of size 512, and a random feature minibatch of size 128.