A Caltech Library Service

Deep Learning in Unconventional Domains


Cvitkovic, Michael William (Milan0 (2020) Deep Learning in Unconventional Domains. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/v7dm-6r52.


Machine learning methods have dramatically improved in recent years thanks to advances in deep learning (LeCun et al., 2015), a set of methods for training high-dimensional, highly-parameterized, nonlinear functions. Yet deep learning progress has been concentrated in the domains of computer vision, vision-based reinforcement learning, and natural language processing. This dissertation is an attempt to extend deep learning into domains where it has thus far had little impact or has never been applied. It presents new deep learning algorithms and state-of-the-art results on tasks in the domains of source-code analysis, relational databases, and tabular data.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:Computer Science; Machine Learning; Deep Learning; Graph Neural Networks; Relational Databases
Degree Grantor:California Institute of Technology
Division:Engineering and Applied Science
Major Option:Computing and Mathematical Sciences
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Vidick, Thomas G.
Thesis Committee:
  • Yue, Yisong (chair)
  • Wierman, Adam C.
  • Vidick, Thomas Georges
  • Anandkumar, Anima
Defense Date:10 March 2020
Non-Caltech Author Email:mwcvitkovic (AT)
Record Number:CaltechTHESIS:04082020-095943405
Persistent URL:
Related URLs:
URLURL TypeDescription adapted for Chapter 2. adapted for Chapter 3. adapted for Chapter 4. adapted for Chapter 6.
Cvitkovic, Michael William (Milan00000-0003-4188-452X
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:13669
Deposited By: Michael Cvitkovic
Deposited On:17 Apr 2020 00:15
Last Modified:17 Jun 2020 19:51

Thesis Files

PDF - Final Version
See Usage Policy.


Repository Staff Only: item control page