- Self-Organising Stochastic Encoders
The processing of mega-dimensional data, such as images, scales linearly with image size only if fixed size processing windows are used. It would be very useful to be able to automate the process of sizing and interconnecting the processing windows. A stochastic encoder that is an extension of the standard Linde-Buzo-Gray vector quantiser, called a stochastic vector quantiser (SVQ), includes this required behaviour amongst its emergent properties, because it automatically splits the input space into statistically independent subspaces, which it then separately encodes. Various optimal SVQs have been obtained, both analytically and numerically. Analytic solutions which demonstrate how the input space is split into independent subspaces may be obtained when an SVQ is used to encode data that lives on a 2-torus (e.g. the superposition of a pair of uncorrelated sinusoids). Many numerical solutions have also been obtained, using both SVQs and chains of linked SVQs: (1) images of multiple independent targets (encoders for single targets emerge), (2) images of multiple correlated targets (various types of encoder for single and multiple targets emerge), (3) superpositions of various waveforms (encoders for the separate waveforms emerge - this is a type of independent component analysis (ICA)), (4) maternal and foetal ECGs (another example of ICA), (5) images of textures (orientation maps and dominance stripes emerge). Overall, SVQs exhibit a rich variety of self-organising behaviour, which effectively discovers the internal structure of the training data. This should have an immediate impact on "intelligent" computation, because it reduces the need for expert human intervention in the design of data processing algorithms.
- A Self-Organising Neural Network for Processing Data from Multiple Sensors
This paper shows how a folded Markov chain network can be applied to the problem of processing data from multiple sensors, with an emphasis on the special case of 2 sensors. It is necessary to design the network so that it can transform a high dimensional input vector into a posterior probability, for which purpose the partitioned mixture distribution network is ideally suited. The underlying theory is presented in detail, and a simple numerical simulation is given that shows the emergence of ocular dominance stripes.
Wednesday, December 22, 2010
Here are some more unpublished papers that I have uploaded to the arXiv:
Posted by Stephen Luttrell at 9:18 am
Friday, December 17, 2010
I have decided to upload some of my unpublished work to the arXiv:
- Adaptive Cluster Expansion (ACE): A Multilayer Network for Estimating Probability Density Functions
We derive an adaptive hierarchical method of estimating high dimensional probability density functions. We call this method of density estimation the "adaptive cluster expansion" or ACE for short. We present an application of this approach, based on a multilayer topographic mapping network, that adaptively estimates the joint probability density function of the pixel values of an image, and presents this result as a "probability image". We apply this to the problem of identifying statistically anomalous regions in otherwise statistically homogeneous images.
- Stochastic Vector Quantisers
In this paper a stochastic generalisation of the standard Linde-Buzo-Gray (LBG) approach to vector quantiser (VQ) design is presented, in which the encoder is implemented as the sampling of a vector of code indices from a probability distribution derived from the input vector, and the decoder is implemented as a superposition of reconstruction vectors, and the stochastic VQ is optimised using a minimum mean Euclidean reconstruction distortion criterion, as in the LBG case. Numerical simulations are used to demonstrate how this leads to self-organisation of the stochastic VQ, where different stochastically sampled code indices become associated with different input subspaces. This property may be used to automate the process of splitting high-dimensional input vectors into low-dimensional blocks before encoding them.
- The Development of Dominance Stripes and Orientation Maps in a Self-Organising Visual Cortex Network (VICON)
A self-organising neural network is presented that is based on a rigorous Bayesian analysis of the information contained in individual neural firing events. This leads to a visual cortex network (VICON) that has many of the properties emerge when a mammalian visual cortex is exposed to data arriving from two imaging sensors (i.e. the two retinae), such as dominance stripes and orientation maps.
Posted by Stephen Luttrell at 1:20 am