Inlämning av Examensarbete / Submission of Thesis

Fahad Khalid MCS-2008:22, pp. 35. TEK/avd. för interaktion och systemdesign, 2008.

The work

Författare / Author: Fahad Khalid
Titel / Title: Measure-based Learning Algorithms: An Analysis of Back-propagated Neural Networks
Abstrakt Abstract:

In this thesis we present a theoretical investigation
of the feasibility of using a problem specific inductive
bias for back-propagated neural networks. We argue
that if a learning algorithm is biased towards optimizing a certain performance measure, it is plausible to assume that it will generate a higher performance score when evaluated using that particular measure. We use the term measure function for a multi-criteria evaluation function that can also be used as an inherent function in learning algorithms, in order to customize the bias of a learning algorithm for a specific problem. Hence, the term measure-based learning algorithms.

We discuss different characteristics of the most
commonly used performance measures and establish
similarities among them. The characteristics of
individual measures and the established similarities are then correlated to the characteristics of the backpropagation algorithm, in order to explore the
applicability of introducing a measure function to backpropagated neural networks.

Our study shows that there are certain
characteristics of the error back-propagation mechanism and the inherent gradient search method that limit the set of measures that can be used for the measure function. Also, we highlight the significance of taking the representational bias of the neural network into account when developing methods for measure-based learning.

The overall analysis of the research shows that
measure-based learning is a promising area of research with potential for further exploration. We suggest directions for future research that might help realize measure-based neural networks.

Populärvetenskaplig beskrivning / Popular science summary: The study is an investigation on the feasibility of using a generic inductive bias for backpropagation artificial neural networks, which could incorporate any one or a
combination of problem specific performance metrics to be optimized. We have identified several limitations of both the standard error backpropagation mechanism as well the inherent gradient search approach. These limitations suggest exploration of methods other than backpropagation, as well use of global search methods instead of gradient search.
Also, we emphasize the importance of taking the representational bias of the neural network in consideration, since only a combination of both procedural and representational bias can provide highly optimal solutions.
Ämnesord / Subject: Datavetenskap - Computer Science\Artificial Intelligence

Nyckelord / Keywords: Supervised learning, Inductive Bias, Artificial Neural Networks

Publication info

Dokument id / Document id:
Program:/ Programme Magisterprogram i Datavetenskap/MSC in Computer science
Registreringsdatum / Date of registration: 06/24/2008
Uppsatstyp / Type of thesis: D-Uppsats/Magister/Master


Handledare / Supervisor: Niklas Lavesson
Examinator / Examiner: Guohua Bai
Organisation / Organisation: Blekinge Institute of Technology
Institution / School: TEK/avd. för interaktion och systemdesign
S-372 25 Ronneby
+46 455 38 50 00

Files & Access

Bifogad uppsats fil(er) / Files attached: measure_based_learning_algorithms.pdf (250 kB, öppnas i nytt fönster)