Machine Learning

@inproceedings{Marsland2014MachineL,
  title={Machine Learning},
  author={Stephen R. Marsland},
  booktitle={Encyclopedia of Social Network Analysis and Mining},
  year={2014}
}
  • S. Marsland
  • Published in
    Encyclopedia of Social…
    2014
  • Computer Science
• Access online or download to your smartphone, tablet or PC/Mac • Search the full text of this and other titles you own • Make and share notes and highlights • Copy and paste text and figures for use in your own documents • Customize your view by changing font size and layout WITH VITALSOURCE® EBOOK ec o n d ed it io n Machine Learning: An Algorithmic Perspective, Second Edition helps you understand the algorithms of machine learning. It puts you on a path toward mastering the relevant… 

A Novel Colour-Constancy Algorithm: A Mixture of Existing Algorithms

TLDR
This paper investigates well-known and state-ofthe-art colour-constancy algorithms and combines them using a weighted-sum approach, showing the proposed method outperforms individual algorithms.

Estimating pore fluid saturation in an oil sands reservoir using ensemble tree machine learning algorithms

This thesis aims to estimate pore fluid saturation values in an oil sands reservoir using ensemble tree based machine learning models. Oil sands reservoirs provide an interesting opportunity to

Use of multiple attractor cellular automata for dimension reduction

TLDR
Properties of rules in rule vector, characterization of attractor's state in Multiple Attractor Cellular Automata works on more than one cell and has two states, three neighbors named elementary cellular automata are studied, and dimension reduction potential of the model is explained.

Using Ultra Wide Band Impulse Radar in Simultaneous Localization and Mapping

TLDR
This thesis investigates if a mobile robot can map and safely navigate in an unknown indoor environment by using Simultaneous Localization and Mapping (SLAM) relying on a distance sensor based on the UWBIR technology for ranging, and shows how features can be extracted from the raw data and combined with positional data to generate maps of the robot's environment, as well as produce through-wall images.