Dr Nicola L C Talbot

I am a chartered mathematician and computer programmer with experience typesetting using LaTeX (pronounced "lay-tek" or "lah-tek"). I am a member of the Institute of Mathematics and its Applications, the East Anglian Writers, and the UK TeX User Group. I'm an honorary lecturer at the School of Computing Sciences, University of East Anglia and am married to Dr Gavin Cawley. (Just to clear up any confusion over my name, my married name is "Mrs Nicola Cawley" and my professional name is "Dr Nicola Talbot".)

Qualifications

B.Sc. (Hons) Class I in Mathematics at the University of Essex, 1991.
Ph.D. entitled "An application-oriented comparison of optimisation and neural-based design techniques". Department of Electronic Systems Engineering, University of Essex, 1996.
Diploma in Creative Writing. University of East Anglia, 2011.

Interviews

Books

LaTeX for Complete Novices Nicola L. C. Talbot. Dickimaw Books, 2012. ISBN 978-1-909440-00-5
Using LaTeX to Write a PhD Thesis Nicola L. C. Talbot. Dickimaw Books, 2013. ISBN 978-1-909440-02-9
LaTeX for Administrative Work Nicola L. C. Talbot. Dickimaw Books, 2015. ISBN 978-1-909440-07-4
The Private Enemy Nicola L. C. Talbot. Dickimaw Books, 2014. ISBN 978-1-909440-05-0
The Foolish Hedgehog Written by Nicola L. C. Talbot, illustrated by Magdalene Pritchett. Dickimaw Books, 2012. ISBN 978-1-909440-01-2
Quack, Quack, Quack. Give My Hat Back! Written by Nicola L. C. Talbot, illustrated by Magdalene Pritchett. Dickimaw Books, 2013. ISBN 978-1-909440-03-6
Cuac, Cuac, Cuac. ¡Devuélveme mi sombrero ya! Written by Nicola L. C. Talbot, illustrated by Magdalene Pritchett, translated by Gonzalo Medina. Dickimaw Books, 2013. ISBN 978-1-909440-06-7

Full Publication List

Talbot NLC and Massara RE (1993), "An application oriented comparison of optimization and neural network based design techniques", In 36th Mid-West Symposium on Circuits and Systems. August, 1993, pp. 261-264.

[BibTeX] [DOI]

BibTeX:

@inproceedings{Talbot1993,
  author = {Nicola L. C. Talbot and Rob E. Massara},
  title = {An application oriented comparison of optimization and neural network based design techniques},
  booktitle = {36th Mid-West Symposium on Circuits and Systems},
  year = {1993},
  pages = {261--264},
  doi = {10.1109/MWSCAS.1993.343080}
}

Barker GC, Talbot NLC and Peck MW (1996), "Risk assessment for microbial contamination hazards: a network approach". November, 1996.

[BibTeX]

BibTeX:

@misc{Barker1996,
  author = {Gary C. Barker and Nicola L. C. Talbot and Mike W. Peck},
  title = {Risk assessment for microbial contamination hazards: a network approach},
  year = {1996},
  note = {Poster presentation and abstract at MAFF Hygienic Food Processing Workshop for Engineers and Microbiologists}
}

Cawley GC and Talbot NLC (1996), "A Fast Index Assignment Algorithm for Vector Quantization over Noisy Transmission Channels", Electronics Letters. July, 1996. Vol. 32(15), pp. 1343-1344.

[BibTeX]

BibTeX:

@article{Cawley1996b,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {A Fast Index Assignment Algorithm for Vector Quantization over Noisy Transmission Channels},
  journal = {Electronics Letters},
  year = {1996},
  volume = {32},
  number = {15},
  pages = {1343--1344}
}

Talbot NLC (1996), "An application-oriented comparison of optimisation and neural-based design techniques". Thesis at: Department of Electronic Systems Engineering, University of Essex. March, 1996.

[BibTeX]

BibTeX:

@phdthesis{Talbot1996b,
  author = {Nicola L. C. Talbot},
  title = {An application-oriented comparison of optimisation and neural-based design techniques},
  school = {Department of Electronic Systems Engineering, University of Essex},
  year = {1996}
}

Talbot NLC and Cawley GC (1996), "A Quadratic Index Assignment Algorithm for Vector Quantisation over Noisy Transmission Channels", In Proceedings of the Institute of Acoustics Autumn Conference (Speech and Hearing 96). November, 1996. Vol. 18, pp. 195-199.

[BibTeX]

BibTeX:

@inproceedings{Talbot1996,
  author = {Talbot, Nicola L. C. and Cawley, Gavin C.},
  title = {A Quadratic Index Assignment Algorithm for Vector Quantisation over Noisy Transmission Channels},
  booktitle = {Proceedings of the Institute of Acoustics Autumn Conference (Speech and Hearing 96)},
  year = {1996},
  volume = {18},
  pages = {195--199}
}

Barker GC, Talbot NLC and Peck MW (1997), "Mathematical aspects of quantitative risk assessment". March, 1997.

[BibTeX]

BibTeX:

@misc{Barker1997a,
  author = {Gary C. Barker and Nicola L. C. Talbot and Mike W. Peck},
  title = {Mathematical aspects of quantitative risk assessment},
  year = {1997},
  note = {Invited oral presentation at BBSRC Food Directorate Microbiology Workshop}
}

Barker GC, Talbot NLC and Peck MW (1997), "Microbial risk assessment: a network approach". January, 1997.

[BibTeX]

BibTeX:

@misc{Barker1997b,
  author = {Gary C. Barker and Nicola L. C. Talbot and Mike W. Peck},
  title = {Microbial risk assessment: a network approach},
  year = {1997},
  note = {Poster presentation and abstract at Department of Health/ACDP Seminar on Microbial Risk Assessment}
}

Talbot NLC and Cawley GC (1997), "A Fast Index Assignment Method for Robust Vector Quantisation of Image Data", In Proceedings of the IEEE International Conference on Image Processing (ICIP-97). Santa Barbara, California, U.S.A. October 26--29, 1997. Vol. 3, pp. 674-677.

[BibTeX]

BibTeX:

@inproceedings{Talbot1997,
  author = {Talbot, Nicola L. C. and Cawley, Gavin C.},
  title = {A Fast Index Assignment Method for Robust Vector Quantisation of Image Data},
  booktitle = {Proceedings of the IEEE International Conference on Image Processing (ICIP-97)},
  year = {1997},
  volume = {3},
  pages = {674--677}
}

Talbot NLC and Massara RE (1997), "A quadratic assignment algorithm that takes module size into account", IEE Electronic Letters.

[BibTeX]

BibTeX:

@article{Talbot1997b,
  author = {Nicola L. C. Talbot and Rob E. Massara},
  title = {A quadratic assignment algorithm that takes module size into account},
  journal = {IEE Electronic Letters},
  year = {1997}
}

Peck MW, Talbot NLC and Barker GC (1998), "Quantitative risk assessment for Clostridium botulinum in minimally processed foods". November, 1998.

[BibTeX]

BibTeX:

@misc{Peck1998,
  author = {Mike W. Peck and Nicola L. C. Talbot and Gary C. Barker},
  title = {Quantitative risk assessment for Clostridium botulinum in minimally processed foods},
  year = {1998},
  note = {Oral presentation and abstract at IBRCC (Interagency botulism research co-ordinating committee) meeting}
}

Barker GC, Talbot NLC and Peck MW (1999), "Microbial risk assessment for sous-vide foods", Third European Symposium on Sous-Vide. Leuvan, March, 1999.

[BibTeX]

BibTeX:

@article{Barker1999a,
  author = {Gary C. Barker and Nicola L. C. Talbot and Mike W. Peck},
  title = {Microbial risk assessment for sous-vide foods},
  journal = {Third European Symposium on Sous-Vide},
  year = {1999}
}

Peck MW, Talbot NLC and Barker GC (1999), "Risk assessment for spore-forming bacteria in food: Bayesian belief representations", In Food Microbiology and Food Safety into the next millennium. The Netherlands, pp. 442-443. Foundation Food Micro '99: A.J.Zeist.

[BibTeX]

BibTeX:

@inproceedings{Peck1999,
  author = {Mike W. Peck and Nicola L. C. Talbot and Gary C. Barker},
  editor = {A. C. J. Tuitelaars and R. A. Samson and F. M. Rombouts and S. Notermans},
  title = {Risk assessment for spore-forming bacteria in food: Bayesian belief representations},
  booktitle = {Food Microbiology and Food Safety into the next millennium},
  publisher = {Foundation Food Micro '99: A.J.Zeist},
  year = {1999},
  pages = {442--443}
}

Cawley GC and Talbot NLC (2001), "Manipulation of prior probabilities in support vector classification", In Proceedings of the IEEE/INNS International Joint Conference on Neural Networks (IJCNN-2001). Washington, D.C., U.S.A. July 15--19, 2001, pp. 2433-2438.

[Abstract] [BibTeX]

Abstract: Asymmetric margin error costs for positive and negative examples are
often cited as an efficient heuristic compensating for unrepresentative
priors in training support vector classifiers. In this paper we show
that this heuristic is well justified via simple resampling ideas
applied to the dual Lagrangian defining the 1-norm soft-margin support
vector machine. This observation also provides a simple expression
for the asymptotically optimal ratio of margin error penalties, eliminating
the need for the trial-and-error experimentation normally encountered.
This method allows the use of a smaller, balanced training data set
in problems characterised by widely disparate prior probabilities,
reducing training time. We demonstrate the usefulness of this method
on a real world benchmark problem, that of predicting forest cover
type given only cartographic data

BibTeX:

@inproceedings{Cawley2001d,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {Manipulation of prior probabilities in support vector classification},
  booktitle = {Proceedings of the IEEE/INNS International Joint Conference on Neural Networks (IJCNN-2001)},
  year = {2001},
  pages = {2433--2438}
}

Barker GC, Talbot NLC and Peck MW (2002), "Risk assessment for Clostridium botulinum: a network approach", International Journal of Biodeterioration and Biodegredation. Vol. 50, pp. 167-175.

[Abstract] [BibTeX] [DOI]

Abstract: The construction and implementation of a mathematical framework for
the representation of the hazards that arise from Clostridium
botulinum growth, and toxin production, in food are described. Botulism
has been recognised as a serious foodborne illness for over a century
and, more recently, has become the subject of increased concern due
to changing processing and consumption patterns associated with foods.
In this respect quantitative risk assessment has an increasingly
important role to play in assisting risk management and ensuring
the safety of minimally processed foods and foods with extended shelf
life.DTLpar Bayesian Belief Networks are a type of expert system
that integrates a graphical flow diagram like, representation of
a hazard domain with a powerful technique for combining probabilities.
This technique facilitates the accumulation of understanding and
experience, for particular hazard domains, into computer tools that
can be used to inspect risks and account for decisions.DTLpar Analysis
of the hazards associated with foodborne botulism involves Belief
Network components that represent contamination processes, thermal
death kinetics for spores, germination and growth of cells, toxin
production and patterns of consumer behaviour, etc. These developments
are discussed and three important aspects of the food safety information
supply, complexity, dependency and uncertainty highlighted. The benefits
associated with a Bayesian view of food safety assessment are illustrated
by a Belief Network representation which supports, and prioritises,
decisions and actions that (a) minimise the chances and extent of
the detrimental events and (b) maximise opportunities for awareness
and control.

BibTeX:

@article{Barker2002,
  author = {Gary C. Barker and Nicola L. C. Talbot and Michael W. Peck},
  title = {Risk assessment for Clostridium botulinum: a network approach},
  journal = {International Journal of Biodeterioration and Biodegredation},
  year = {2002},
  volume = {50},
  pages = {167--175},
  doi = {10.1016/S0964-8305(02)00083-5}
}

Cawley GC and Talbot NLC (2002), "A Greedy Training Algorithm for Sparse Least-Squares Support Vector Machines", In Proceedings of the International Conference on Artificial Neural Networks (ICANN-2002). Madrid, Spain, August 27--30, 2002. Vol. 2415, pp. 681-686. Springer.

[Abstract] [BibTeX]

Abstract: Suykens et al. describes a form of kernel ridge regression
known as the least-squares support vector machine (LS-SVM). In this
paper, we present a simple, but efficient, greedy algorithm for constructing
near optimal sparse approximations of least-squares support vector
machines, in which at each iteration the training pattern minimising
the regularised empirical risk is introduced into the kernel expansion.
The proposed method demonstrates superior performance when compared
with the pruning technique described by Suykens et al., over
the motorcycle and Boston housing datasets

BibTeX:

@inproceedings{Cawley2002c,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {A Greedy Training Algorithm for Sparse Least-Squares Support Vector Machines},
  booktitle = {Proceedings of the International Conference on Artificial Neural Networks (ICANN-2002)},
  publisher = {Springer},
  year = {2002},
  volume = {2415},
  pages = {681--686}
}

Cawley GC and Talbot NLC (2002), "Efficient formation of a basis in a kernel induced feature space", In Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2002). Bruges, Belgium, April 24--26, 2002, pp. 1-6.

[Abstract] [BibTeX]

Abstract: Baudat and Anouar (2001) propose a simple greedy algorithm for estimation
of an approximate basis of the subspace spanned by a set of fixed
vectors embedded in a kernel induced feature space. The resulting
set of basis vectors can then be used to construct sparse kernel
expansions for classification and regression tasks. In this paper
we describe five algorithmic improvements to the method of Baudat
and Anouar, allowing the construction of an approximate basis with
a computational complexity that is independent of the number of training
patterns, depending only on the number of basis vectors extracted

BibTeX:

@inproceedings{Cawley2002a,
  author = {Cawley, G. C. and Talbot, N. L. C.},
  title = {Efficient formation of a basis in a kernel induced feature space},
  booktitle = {Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2002)},
  year = {2002},
  pages = {1--6}
}

Cawley GC and Talbot NLC (2002), "Improved Sparse Least-Squares Support Vector Machines", Neurocomputing. October, 2002. Vol. 48, pp. 1025-1031.

[Abstract] [BibTeX] [DOI]

Abstract: Suykens et al. describe a weighted least-squares formulation
of the support vector machine for regression problems and presents
a simple algorithm for sparse approximation of the typically fully
dense kernel expansions obtained using this method. In this paper,
we present an improved method for achieving sparsity in least-squares
support vector machines, which takes into account the residuals for
all training patterns, rather than only those incorporated in the
sparse kernel expansion. The superiority of this algorithm is demonstrated
on the motorcycle and Boston housing datasets

BibTeX:

@article{Cawley2002b,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {Improved Sparse Least-Squares Support Vector Machines},
  journal = {Neurocomputing},
  year = {2002},
  volume = {48},
  pages = {1025--1031},
  doi = {10.1016/S0925-2312(02)00606-9}
}

Cawley GC and Talbot NLC (2002), "Reduced rank kernel ridge regression", Neural Processing Letters. December, 2002. Vol. 16(3), pp. 293-302.

[Abstract] [BibTeX] [DOI]

Abstract: Ridge regression is a classical statistical technique that attempts
to address the bias-variance trade-off in the design of linear regression
models. A reformulation of ridge regression in dual variables permits
a non-linear form of ridge regression via the well-known ``kernel
trick''. Unfortunately, unlike support vector regression models,
the resulting kernel expansion is typically fully dense. In this
paper, we introduce a reduced rank kernel ridge regression (RRKRR)
algorithm, capable of generating an optimally sparse kernel expansion
that is functionally identical to that resulting from conventional
kernel ridge regression (KRR). The proposed method is demonstrated
to out-perform an alternative sparse kernel ridge regression algorithm
on the Motorcycle and Boston Housing benchmarks

BibTeX:

@article{Cawley2002f,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {Reduced rank kernel ridge regression},
  journal = {Neural Processing Letters},
  year = {2002},
  volume = {16},
  number = {3},
  pages = {293--302},
  doi = {10.1023/A:1021798002258}
}

Foxall RJ, Cawley GC, Talbot NLC, Dorling SR and Mandic DP (2002), "Heteroscedastic regularised kernel regression for prediction of episodes of poor air quality", In Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2002). Bruges, Belgium, April 24--26, 2002, pp. 19-24.

[Abstract] [BibTeX]

Abstract: A regularised kernel regression model is introduced for data characterised
by a heteroscedastic (input dependent variance) Gaussian noise process.
The proposed model provides more robust estimates of the conditional
mean than standard models and also confidence intervals (error bars)
on predictions. The benefits of the proposed model are demonstrated
for the task of non-linear prediction of episodes of poor air quality
in urban environments

BibTeX:

@inproceedings{Foxall2002a,
  author = {Foxall, Robert J. and Cawley, Gavin C. and Talbot, Nicola L. C. and Dorling, Stephen R. and Mandic, Danilo P.},
  title = {Heteroscedastic regularised kernel regression for prediction of episodes of poor air quality},
  booktitle = {Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2002)},
  year = {2002},
  pages = {19--24}
}

Saadi K, Cawley GC and Talbot NLC (2002), "Fast exact leave-one-out cross-validation of least-squares support vector machines", In Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2002). Bruges, Belgium, April 24--26, 2002, pp. 149-154.

[Abstract] [BibTeX]

Abstract: Model selection methods for kernel machines often seek to minimise
an upper bound on the leave-one-out cross-validation error. This
paper describes an efficient algorithm for exact leave-one-out
cross-validation of least-squares support vector machines, in both
classification and regression settings. The proposed method exploits
the considerable redundancy in the family of systems of linear equations
to be solved in explicit computation of the leave-one-out error.
The efficiency of the proposed approach is demonstrated using real-world
and synthetic benchmark datasets

BibTeX:

@inproceedings{Saadi2002,
  author = {Saadi, K. and Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {Fast exact leave-one-out cross-validation of least-squares support vector machines},
  booktitle = {Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2002)},
  year = {2002},
  pages = {149--154}
}

Cawley GC and Talbot NLC (2003), "Efficient cross-validation of kernel Fisher discriminant classifiers", In Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2003). Bruges, Belgium, April 23-25, 2003, pp. 241-246.

[BibTeX]

BibTeX:

@inproceedings{Cawley2003a,
  author = {Gavin C. Cawley and Nicola L. C. Talbot},
  title = {Efficient cross-validation of kernel Fisher discriminant classifiers},
  booktitle = {Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2003)},
  year = {2003},
  pages = {241--246}
}

Cawley GC and Talbot NLC (2003), "Efficient leave-one-out cross-validation of kernel Fisher discriminant classifiers", Pattern Recognition. November, 2003. Vol. 36(11), pp. 2585-2592.

[BibTeX] [DOI]

BibTeX:

@article{Cawley2003d,
  author = {Gavin C. Cawley and Nicola L. C. Talbot},
  title = {Efficient leave-one-out cross-validation of kernel Fisher discriminant classifiers},
  journal = {Pattern Recognition},
  year = {2003},
  volume = {36},
  number = {11},
  pages = {2585--2592},
  doi = {10.1016/S0031-3203(03)00136-5}
}

Cawley GC, Talbot NLC, Foxall RJ, Dorling SR and Mandic DP (2003), "Unbiased Estimation of Conditional Variance in Heteroscedastic Kernel Ridge Regression", In Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2003). Bruges, Belgium, April 23--25, 2003, pp. 209-214.

[BibTeX]

BibTeX:

@inproceedings{Cawley2003b,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C. and Foxall, Robert J. and Dorling, Stephen R. and Mandic, Danilo P.},
  title = {Unbiased Estimation of Conditional Variance in Heteroscedastic Kernel Ridge Regression},
  booktitle = {Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2003)},
  year = {2003},
  pages = {209--214}
}

Cawley GC and Talbot NLC (2004), "Efficient model selection for kernel logistic regression", In Proceedings of the 17th International Conference on Pattern Recognition (ICPR-2004). Cambridge, United Kingdom, August 23--26, 2004. Vol. 2, pp. 439-442.

[Abstract] [BibTeX] [DOI]

Abstract: Kernel logistic regression models, like their linear counterparts,
can be trained using the efficient iteratively re-weighted least-squares
(IRWLS) algorithm. This approach suggests an approximate leave-one-out
cross-validation estimator based on an existing method for exact
leave-one-out cross-validation of least-squares models. Results compiled
over seven benchmark datasets are presented for kernel logistic regression
with model selection procedures based on both conventional k-fold
and approximate leave-one-out cross-validation criteria, demonstrating
the proposed approach to be viable.

BibTeX:

@inproceedings{Cawley2004a,
  author = {Gavin C. Cawley and Nicola L. C. Talbot},
  title = {Efficient model selection for kernel logistic regression},
  booktitle = {Proceedings of the 17th International Conference on Pattern Recognition (ICPR-2004)},
  year = {2004},
  volume = {2},
  pages = {439--442},
  doi = {10.1109/ICPR.2004.1334249}
}

Cawley GC and Talbot NLC (2004), "Fast leave-one-out cross-validation of sparse least-squares support vector machines", Neural Networks. December, 2004. Vol. 17(10), pp. 1467-1475.

[BibTeX] [DOI]

BibTeX:

@article{Cawley2004c,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {Fast leave-one-out cross-validation of sparse least-squares support vector machines},
  journal = {Neural Networks},
  year = {2004},
  volume = {17},
  number = {10},
  pages = {1467--1475},
  doi = {10.1016/j.neunet.2004.07.002}
}

Cawley GC and Talbot NLC (2004), "Sparse Bayesian kernel logistic regression", In Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2004). Bruges, Belgium, April 28--30, 2004, pp. 133-138.

[BibTeX]

BibTeX:

@inproceedings{Cawley2004b,
  author = {Gavin C. Cawley and Nicola L. C. Talbot},
  title = {Sparse Bayesian kernel logistic regression},
  booktitle = {Proceedings of the European Symposium on Artificial Neural Networks (ESANN-2004)},
  year = {2004},
  pages = {133--138}
}

Cawley GC and Talbot NLC (2004), "The Evidence Framework applied to Sparse Kernel Logistic Regression", Neurocomputing. Vol. 64 (Trends in Neurocomputing : 12th European Symposium on Artificial Neural Networks 2004), pp. 119-135.

[BibTeX] [DOI]

BibTeX:

@article{Cawley2004d,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {The Evidence Framework applied to Sparse Kernel Logistic Regression},
  journal = {Neurocomputing},
  year = {2004},
  volume = {64 (Trends in Neurocomputing : 12th European Symposium on Artificial Neural Networks 2004)},
  pages = {119--135},
  doi = {10.1016/j.neucom.2004.11.021}
}

Cawley GC, Talbot NLC, Foxall RJ, Dorling SR and Mandic DP (2004), "Heteroscedastic kernel ridge regression", Neurocomputing. March, 2004. Vol. 57, pp. 105-124.

[Abstract] [BibTeX] [DOI]

Abstract: In this paper we extend a form of kernel ridge regression (KRR) for
data characterised by a heteroscedastic (i.e. input dependent variance)
Gaussian noise process, introduced in Foxall et al. Foxall2002a.
It is shown that the proposed heteroscedastic kernel ridge regression
model can give a more accurate estimate of the conditional mean of
the target distribution than conventional KRR and also provides an
indication of the spread of the target distribution (i.e. predictive
error bars). The leave-one-out cross-validation estimate of the conditional
mean is used in fitting the model of the conditional variance in
order to overcome the inherent bias in maximum likelihood estimates
of the variance. The benefits of the proposed model are demonstrated
on synthetic and real-world benchmark data sets and for the task
of predicting episodes of poor air quality in an urban environment.

BibTeX:

@article{Cawley2004e,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C. and Foxall, Robert J. and Dorling, Stephen R. and Mandic, Danilo P.},
  title = {Heteroscedastic kernel ridge regression},
  journal = {Neurocomputing},
  year = {2004},
  volume = {57},
  pages = {105--124},
  doi = {10.1016/j.neucom.2004.01.005}
}

Saadi K, Talbot NLC and Cawley GC (2004), "Optimally regularised kernel Fisher discriminant analysis", In Proceedings of the 17th International Conference on Pattern Recognition (ICPR-2004). Cambridge, United Kingdom, August 23--26, 2004. Vol. 2, pp. 427-430.

[BibTeX] [DOI]

BibTeX:

@inproceedings{Saadi2004a,
  author = {K. Saadi and Nicola L. C. Talbot and Gavin C. Cawley},
  title = {Optimally regularised kernel Fisher discriminant analysis},
  booktitle = {Proceedings of the 17th International Conference on Pattern Recognition (ICPR-2004)},
  year = {2004},
  volume = {2},
  pages = {427--430},
  doi = {10.1109/ICPR.2004.1334245}
}

Cawley G, Talbot N, Janacek G and Peck M (2005), "Bayesian Kernel Learning Methods for Parametric Accelerated Life Survival Analysis", In Deterministic and Statistical Methods in Machine Learning. Vol. 3635, pp. 37-55. Springer Berlin / Heidelberg.

[BibTeX] [DOI]

BibTeX:

@incollection{Cawley2005b,
  author = {Cawley, Gavin and Talbot, Nicola and Janacek, Gareth and Peck, Michael},
  editor = {Winkler, Joab and Niranjan, Mahesan and Lawrence, Neil},
  title = {Bayesian Kernel Learning Methods for Parametric Accelerated Life Survival Analysis},
  booktitle = {Deterministic and Statistical Methods in Machine Learning},
  publisher = {Springer Berlin / Heidelberg},
  year = {2005},
  volume = {3635},
  pages = {37-55},
  doi = {10.1007/11559887_3}
}

Cawley GC and Talbot NLC (2005), "A simple trick for constructing Bayesian formulations of sparse kernel learning methods", In Proceedings of the IEEE/INNS International Joint Conference on Neural Networks (IJCNN-2005). Montreal, Canada, July 31 -- August 4, 2005, pp. 1425-1430.

[BibTeX]

BibTeX:

@inproceedings{Cawley2005d,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {A simple trick for constructing Bayesian formulations of sparse kernel learning methods},
  booktitle = {Proceedings of the IEEE/INNS International Joint Conference on Neural Networks (IJCNN-2005)},
  year = {2005},
  pages = {1425--1430}
}

Cawley GC and Talbot NLC (2005), "Constructing Bayesian formulations of sparse kernel learning methods", Neural Networks. July--August, 2005. Vol. 18, issues 5--6, pp. 674-683.

[BibTeX] [DOI]

BibTeX:

@article{Cawley2005e,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {Constructing Bayesian formulations of sparse kernel learning methods},
  journal = {Neural Networks},
  year = {2005},
  volume = {18, issues 5--6},
  pages = {674--683},
  doi = {10.1016/j.neunet.2005.06.002}
}

Cawley GC and Talbot NLC (2005), "Sparse Bayesian learning and the relevance multi-layer perceptron network", In Proceedings of the IEEE/INNS International Joint Conference on Neural Networks (IJCNN-2005). Montreal, Canada, July 31 -- August 4, 2005, pp. 1320-1324.

[BibTeX]

BibTeX:

@inproceedings{Cawley2005c,
  author = {Cawley, Gavin C. and Talbot, Nicola L. C.},
  title = {Sparse Bayesian learning and the relevance multi-layer perceptron network},
  booktitle = {Proceedings of the IEEE/INNS International Joint Conference on Neural Networks (IJCNN-2005)},
  year = {2005},
  pages = {1320--1324}
}

Cawley GC and Talbot NLC (2005), "The evidence framework applied to sparse kernel logistic regression", Neurocomputing. Vol. 64, pp. 119-135.

[BibTeX] [DOI]

BibTeX:

@article{Cawley2005f,
  author = {Gavin C. Cawley and Nicola L. C. Talbot},
  title = {The evidence framework applied to sparse kernel logistic regression},
  journal = {Neurocomputing},
  year = {2005},
  volume = {64},
  pages = {119--135},
  doi = {10.1016/j.neucom.2004.11.021}
}

Flom P, Hagen H, Hogg J, Talbot N, Taylor P, Thiele C and Walden D (2005), "What is TeX?", The PracTeX Journal. Vol. 3