ML Anthology

Random Projections and Α-Shape to Support the Kernel Design (Student Abstract)

Daniel Moreira Cestari, Rodrigo Fernandes de Mello
AAAI 2020 pp. 13877-13878
doi:10.1609/AAAI.V34I10.7211 /aaai/2020/cestari2020aaai-random/

Abstract

We demonstrate that projecting data points into hyperplanes is good strategy for general-purpose kernel design. We used three different hyperplanes generation schemes, random, convex hull and α-shape, and evaluated the results on two synthetic and three well known image-based datasets. The results showed considerable improvement in the classification performance in almost all scenarios, corroborating the claim that such an approach can be used as a general-purpose kernel transformation. Also, we discuss some connection with Convolutional Neural Networks and how such an approach could be used to understand such networks better.

PDF AAAI Semantic Scholar

Cite

Text

Cestari and de Mello. "Random Projections and Α-Shape to Support the Kernel Design (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020. doi:10.1609/AAAI.V34I10.7211

Markdown

[Cestari and de Mello. "Random Projections and Α-Shape to Support the Kernel Design (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020.](https://mlanthology.org/aaai/2020/cestari2020aaai-random/) doi:10.1609/AAAI.V34I10.7211

BibTeX

@inproceedings{cestari2020aaai-random,
  title     = {{Random Projections and Α-Shape to Support the Kernel Design (Student Abstract)}},
  author    = {Cestari, Daniel Moreira and de Mello, Rodrigo Fernandes},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2020},
  pages     = {13877-13878},
  doi       = {10.1609/AAAI.V34I10.7211},
  url       = {https://mlanthology.org/aaai/2020/cestari2020aaai-random/}
}