Parametric Manifold Learning of Gaussian Mixture Models
Authors: Ziquan Liu, Lei Yu, Janet H. Hsiao, Antoni B. Chan
IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the effectiveness of PML-GMM through experiments on synthetic, eye-fixation, flow cytometry, and social check-in data. |
| Researcher Affiliation | Academia | Ziquan Liu1 , Lei Yu1 , Janet H. Hsiao2 and Antoni B. Chan1 1Department of Computer Science, City University of Hong Kong 2Department of Psychology, The University of Hong Kong |
| Pseudocode | No | The paper describes the algorithm steps in prose (e.g., Variational M-step, Variational E-step) but does not include a structured pseudocode or algorithm block. |
| Open Source Code | Yes | [Liu et al., 2019] Ziquan Liu, Lei Yu, Janet H. Hsiao, and Antoni B. Chan. Parametric manifold learning of Gaussian mixture models-Supplementary. http://visal.cs.cityu.edu.hk/static/pubs/conf/ijcai19-pml-gmm-supp.pdf, 2019. |
| Open Datasets | Yes | The synthetic dataset consists of 39 synthetic GMMs... generated from a 1D latent space... (see supplemental [Liu et al., 2019])... We use an open AML dataset [Aghaeepour et al., 2013]... The eye-fixation data [Chan et al., 2018]... social check-in data [Cho et al., 2011]. |
| Dataset Splits | Yes | For a fair comparison, the latent space dimension is set as 1 for all methods. 20 GMMs are used for training (blue points) and the other 19 (red points) are reserved for testing. ... The eye-fixation data [Chan et al., 2018] consists of eye-fixation coordinates of 34 young adults and 67 older adults (34 for training, 33 for testing) when recognizing faces. ... 30 healthy and 30 unhealthy patients are used for training, and another 10 healthy and 10 unhealthy are used for testing. |
| 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 does not provide specific software dependencies with version numbers used for replication. |
| Experiment Setup | Yes | For PML-GMM, the latent space is set to 6 dimensions (dw = dz = dy = 2) and the HLS is set to dv = 3. We run EM on each patient s data using K {2, 3, 4, 5}. |