Training Recurrent Neural Networks - ProQuest (original) (raw)

At the request of the author, this graduate work is not available to view or purchase.

Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works.

Your library or institution may give you access to the complete full text of other documents in ProQuest that are available as full text.

Once connected, you can find other documents related to your research within ProQuest, as well as cite, email or print them.

Abstract/Details

Recurrent Neural Networks (RNNs) are powerful sequence models that were believed to be difficult to train, and as a result they were rarely used in machine learning applications. This thesis presents methods that overcome the difficulty of training RNNs, and applications of RNNs to challenging problems.

We first describe a new probabilistic sequence model that combines Restricted Boltzmann Machines and RNNs. The new model is more powerful than similar models while being less difficult to train.

Next, we present a new variant of the Hessian-free (HF) optimizer and show that it can train RNNs on tasks that have extreme long-range temporal dependencies, which were previously considered to be impossibly hard. We then apply HF to character-level language modelling and get excellent results.

We also apply HF to optimal control and obtain RNN control laws that can successfully operate under conditions of delayed feedback and unknown disturbances.

Finally, we describe a random parameter initialization scheme that allows gradient descent with momentum to train RNNs on problems with long-term dependencies. This directly contradicts widespread beliefs about the inability of first-order methods to do so, and suggests that previous attempts at training RNNs failed partly due to flaws in the random initialization.

Classification

0984: Computer science

Identifier / keyword

Applied sciences; HF optimizer; Hessian-free optimizer; RNNs; Recurrent Neural Networks; Restricted Boltzmann Machines; Temporal dependencies

Title

Training Recurrent Neural Networks

Source

DAI-B 75/06(E), Dissertation Abstracts International

University/institution

University of Toronto (Canada)

University location

Canada -- Ontario, CA

Source type

Dissertation or Thesis

Document type

Dissertation/Thesis

Dissertation/thesis number

NS22066

ProQuest document ID

1501655550

Copyright

Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works.

Document URL

https://www.proquest.com/docview/1501655550