M-Statistic for Kernel Change-Point Detection
Authors: Shuang Li, Yao Xie, Hanjun Dai, Le Song
NeurIPS 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We show that our methods perform well in both synthetic and real world data. |
| Researcher Affiliation | Academia | Shuang Li, Yao Xie H. Milton Stewart School of Industrial and Systems Engineering Georgian Institute of Technology sli370@gatech.edu yao.xie@isye.gatech.edu Hanjun Dai, Le Song Computational Science and Engineering College of Computing Georgia Institute of Technology hanjundai@gatech.edu lsong@cc.gatech.edu |
| Pseudocode | No | The paper describes the proposed methods in text and mathematical formulas but does not provide structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any explicit statements about releasing source code for the described methodology, nor does it provide a link to a code repository. |
| Open Datasets | Yes | Our datasets include: (1) CENSREC-1-C: a real-world speech dataset in the Speech Resource Consortium (SRC) corpora provided by National Institute of Informatics (NII)1; (2) Human Activity Sensing Consortium (HASC) challenge 2011 data2. 1Available from http://research.nii.ac.jp/src/en/CENSREC-1-C.html 2Available from http://hasc.jp/hc2011 |
| Dataset Splits | No | The paper uses the term 'validation' in the context of validating theoretical results (e.g., 'The numerical accuracy of our approximations are validated by numerical examples.') and does not provide specific train/validation/test dataset splits for reproducibility. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions mathematical concepts like 'Gaussian kernel' and 'median trick' but does not specify any software libraries or dependencies with version numbers used in the experiments. |
| Experiment Setup | Yes | In the following examples, we use a Gaussian kernel: k(Y, Y 0) = exp[-k Y Y 0k2/2σ2], where σ > 0 is the kernel bandwidth and we use the median trick [10, 8] to get the bandwidth which is estimated using the background data. We choose the maximum block size to be Bmax = 20. In the online setting, we compare EDD...when the signal is 20 dimensional and the transition happens from a zero-mean Gaussian N(0, I20) to a non-zero mean Gaussian N(µ, I20), where the postchange mean vector µ is element-wise equal to a constant mean shift. Also, Figure 2 caption mentions N=5. |