Browsing Computer Science by Title
Now showing items 2-21 of 82
-
A comparative study of function approximators involving neural networks
Various researchers have used one hidden layer neural networks (weighted sums of sigmoids) to find the unknown solutions of partial differential equations. They are embedded within an energy function that describes the ... -
A computational platform for simulating reach-to-grasp actions: modelling physics, touch receptors and motor control mechanisms
Reaching to grasp is a fundamental human action. In this thesis, I present a new computational platform for simulating reach-to-grasp actions and exploring how they are learned and performed. There are three innovations ... -
A Mechanism for Attending to 'Unusual' Object Properties, and its Role in a Model of Object Categorization
Categorization is one of the most fundamental aspects of human cognition. From soon after birth, humans form categories based on most of our sensory inputs. Stevan Harnad (2005) famously said, “To cognize is to categorize: ... -
A Natural Interface for 3D Manipulation
3D interfaces enable a user to create and manipulate 3D objects in a computer-simulated 3-dimensional world. There are many existing interfaces, but few of them exhibit the naturalness which characterizes our interaction ... -
A Neural Network Model of Causative Actions
Many of the actions we perform are defined by the effects they bring about, rather than as stereotypical sequences of motor movements. For instance, to open a door we must perform an action which results in the door opening. ... -
A pragmatic tutorial dialogue system: design, implementation and evaluation in a health sciences domain
In large undergraduate classes, it is time-consuming, costly and seldom practical for the teacher to provide individualised feedback to students on their written responses to questions. A request for advice in relation ... -
A Statistical Model of Error Correction for Computer Assisted Language Learning Systems
This thesis presents a study in the area of computer assisted language learning systems. The study focusses on the topic of automatic correction of student errors. In the thesis, I will describe a novel statistical model ... -
Achieving continual learning in deep neural networks through pseudo-rehearsal
Neural networks are very powerful computational models, capable of outperforming humans on a variety of tasks. However, unlike humans, these networks tend to catastrophically forget previous information when learning new ... -
Actions of attention, and attention to action: investigating the relationship between visual attention, episodic representation, and language
Cognitive representations of episodes are likely to play an important role in the neural mechanisms representing the syntactic and semantic structure of natural language sentences. However, there is little consensus on how ... -
An examination of polychronous neural groups as a potential substrate for representation
It is widely assumed that synaptic plasticity provides the neural basis for long-term memory in the brain (Abraham, 2008, Caporale and Dan, 2008, Martin et al., 2000). However, the precise nature of the underlying ... -
An Indoor Localisation System Based on Ubiquitous Technology
Indoor localisation systems are concerned with locating people or devices within an indoor environment. Within the field, there are a large number of technologies available for this task. Depending on the technology and ... -
Approaches to an Efficient and Effective Search Engine
Information retrieval (IR) is the process of finding relevant information, based on user queries, in a large collection of documents. The two main performance issues in IR are effectiveness and efficiency; effectiveness ... -
Calculating the 2D motion of lumbar vertebrae using splines
We present a new method for calculating the transformation parameters for a rigid body undergoing planar motion parallel to the image plane. The method utilises splines to represent the outline of the rigid shape rather ... -
Catastrophic Forgetting in Reinforcement-Learning Environments
Reinforcement learning (RL) problems are a fundamental part of machine learning theory, and neural networks are one of the best known and most successful general tools for solving machine learning problems. Despite this, ... -
Click Logs as a Source of Data for Evaluating the Quality of Hypertext
The automatic generation of useful hyperlinks has been a topic of interest in the field of information retrieval for several years, but no existing method for evaluating the quality of the generated hyperlinks has yet been ... -
Collaborative Learning of Fine-grained Visual Data
Problem: Deep learning based vision systems have achieved near human accuracy in recognizing coarse object categories from visual data. But recognizing fine-grained sub-categories remains an open problem. Tasks like ... -
Communicating agents: an emerging approach for distributed heterogeneous systems
The concept of an intelligent software agent has emerged from its origins in artificial intelligence laboratories to become an important basis for the development of distributed systems in the mainstream computer science ... -
Computational Modelling of Synaptic Plasticity in the Dentate Gyrus Granule Cell
After more than 30 years of study, the dynamics of synaptic plasticity in neurons still remain somewhat a mystery. By conducting a series of simulations on a simulated version of the rat dentate gyrus granule cell usingthe ... -
Convolutional Neural Networks for Medical Image Semantic Segmentation
This thesis focuses on the problem of medical image segmentation using convolutional neural networks (CNN) with a particular focus on solving the issues that arise in boundary regions and high-frequency areas using different ... -
Decentralized Broadcast for Reliable and Real-time Dissemination of Cooperative Awareness Messages in Vehicular Ad-hoc Networks
Reliable broadcast has long been a challenging problem for wireless communications due to unstable link connectivities. Recently, it has become an increasing concern for Vehicular Ad-hoc Networks (VANETs), especially for ...