Comparative Study on CNN Initialization

Supervision: Shadi Albarqouni

Abstract

Tasks:

Compare recent literatures on CNN initialization

Requirements:

Good understanding of statistics and machine learning methods.
Very good programming skills in C++/Python

Location:

FMI Building, Garching

Literature

[1] Krähenbühl, Philipp, et al. "Data-dependent initializations of convolutional neural networks." arXiv preprint arXiv:1511.06856 (2015).
[2] Mishkin, Dmytro, and Jiri Matas. "All you need is a good init." arXiv preprint arXiv:1511.06422 (2015).

ProjectForm
Title:	Comparative Study on CNN Initialization
Abstract:
Student:	Michael Wengler
Director:	Prof. Nassir Navab
Supervisor:	Shadi Albarqouni
Type:	Bachelor Thesis
Area:
Status:	finished
Start:
Finish:
Thesis (optional):
Picture: