Abstract for freitas_tr328

Cambridge University Engineering Department Technical Report CUED/F-INFENG/TR328

SEQUENTIAL MONTE CARLO METHODS FOR OPTIMISATION OF NEURAL NETWORKS MODELS.

JFG de Freitas, M Niranjan, AH Gee and A Doucet

July 1998

We propose a novel strategy for training neural networks using sequential Monte Carlo algorithms and develop an efficient hybrid gradient descent/importance resampling algorithm. This global optimisation strategy allows us to learn the probability distribution of the network weights in a sequential framework. It is well suited to applications involving on-line, nonlinear or non-stationary signal processing. We show how the new algorithms outperform extended Kalman filter training and radial basis function networks on several problems. We address the problem of pricing option contracts, traded in financial markets. In this context, the sequential algorithms are used to estimate the parameters of sequential Black-Scholes models and multi-layer perceptrons. We found that both model structures allowed us to successfully compute the price of options on the FTSE100 index. In addition, we were able to estimate the complete probability density functions of the risk-free interest rate and implied volatility on a daily basis.


(ftp:) freitas_tr328.ps.gz (http:) freitas_tr328.ps.gz
PDF (automatically generated from original PostScript document - may be badly aliased on screen):
  (ftp:) freitas_tr328.pdf | (http:) freitas_tr328.pdf

If you have difficulty viewing files that end '.gz', which are gzip compressed, then you may be able to find tools to uncompress them at the gzip web site.

If you have difficulty viewing files that are in PostScript, (ending '.ps' or '.ps.gz'), then you may be able to find tools to view them at the gsview web site.

We have attempted to provide automatically generated PDF copies of documents for which only PostScript versions have previously been available. These are clearly marked in the database - due to the nature of the automatic conversion process, they are likely to be badly aliased when viewed at default resolution on screen by acroread.