Robust Kernel Density Estimation by Scaling and Projection in Hilbert Space
Authors: Robert A Vandermeulen, Clayton Scott
NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the robustness of the SPKDE with numerical experiments and a consistency result which shows that asymptotically the SPKDE recovers the uncontaminated density under sufficient conditions on the contamination. In this paper we present a new formalism for nonparametric density estimation, necessary and sufficient conditions for decontamination, the construction of the SPKDE, and a proof of consistency. We also include experimental results applying the algorithm to benchmark datasets with comparisons to the RKDE, traditional KDE, and an alternative robust KDE implementation. |
| Researcher Affiliation | Academia | Robert A. Vandermeulen Department of EECS University of Michigan Ann Arbor, MI 48109 rvdm@umich.edu Clayton D. Scott Deparment of EECS Univeristy of Michigan Ann Arbor, MI 48109 clayscot@umich.edu |
| Pseudocode | No | The paper describes algorithms but does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to source code (specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described. |
| Open Datasets | No | The paper mentions using "12 classification datasets (names given in the supplemental material)" and refers to them as "benchmark datasets," but it does not provide specific citations, links, or direct statements of public availability within the main text. |
| Dataset Splits | Yes | For each permutation of each dataset, the bandwidth parameter is set using the training data with a LOOCV line search minimizing DKL fobs|| bf , where bf is the KDE based on the contaminated data and fobs is the observed density. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper mentions optimization methods and refers to a paper describing an algorithm for projections but does not list specific software dependencies with version numbers (e.g., library names with version numbers). |
| Experiment Setup | Yes | For the SPKDE the parameter β was chosen to be 2 for all experiments. This choice of β is based on a few preliminary experiments for which it yielded good results over various sample contamination amounts. The Gaussian kernel was used for all density estimates. |