Hugh Cartwright
Artificial Neural Networks
Artificial Neural
Networks (ANNs) lie at the heart of the burgeoning field of
Deep Learning.
The purpose of an
ANN is to link patterns of input data to associated output
data. For example, pharmaceutical companies generally
wish to investigate possible links between the structure of
potential drugs (molecular structure providing the input) and
their effectiveness in treating disease (the output). Although
it is reasonable to suppose that such correlations do exist,
uncovering those relationships is a challenging task.
ANNs comprise: (i)
a set of computational nodes, each of which is able to perform
simple mathematical operations; (ii) connections between those
nodes; (iii) weights on the connections, and (iv) a
computational recipe which defines how each node will
calculate an output signal from the input signals it receives.
The nodes are
generally arranged in layers, the first of which accepts input
from the outside. The input is passed through each layer of
nodes in turn, with the final layer generating output
signal(s). Learning is accomplished in the network through
adjustment of the connection weights, which form the memory of
the system. This process, known as "training", is generally
the slowest step in using an ANN, often being hundreds or
thousands of times slower than the actual application of the
fully-trained network.
Representative Books:
§§ Hugh M.
Cartwright (ed.) Artificial Neural Networks, (3rd edn.)
A forthcoming text in Springer's "Methods in Molecular
Biology" series. Publication early 2020.
§§ Hugh M.
Cartwright (ed.) Artificial
Neural Networks, (2nd edn.) Vol 1260 in the Methods in
Molecular Biology series. Springer, New York. (2015).
ISBN 978-1-4939-22383-3.
§§ Hugh M.
Cartwright, Using
Artificial Intelligence in Chemistry and Biology: A
Practical Guide, Taylor & Francis, (2008).
§§ Hugh M.
Cartwright & Les M. Sztandera (eds), Soft
Computing Approaches in Chemistry,
Springer-Verlag, Berlin, (2003).
§§ Hugh M.
Cartwright (ed.), Intelligent
Data Analysis in Science, Volume 4 in the Oxford
Chemistry Masters Series, Oxford University Press, (2000).
ISBN 0-19-850233-8.
§§ Hugh M.
Cartwright, Applications
of Artificial Intelligence in Chemistry, Volume 11 in
the Oxford Chemistry Primers Series, Japanese translation,
Maruzen Co, Tokyo, (1995). ISBN 4-621-04088-X
§§ Hugh M.
Cartwright, Applications
of Artificial Intelligence in Chemistry, Volume 11 in
the Oxford Chemistry Primers Series, Oxford University
Press, (1993). ISBN 0-19-855736-1.
Representative Papers:
§ Keith T.
Butler, Daniel W. Davies, Hugh M. Cartwright, Olexandr Isayev
& Aron Walsh. Machine Learning for molecular and materials
science. Nature 559, 547-555 (2018).
§ Hugh M.
Cartwright & Silvia Curteanu, Neural Networks applied in
chemistry. II. Neuro-evolutionary techniques in process
modeling and optimization, Ind & Eng Chem Res,
2013, 52(36), 12673 - 12688. DOI: 10.1021/ie4000954.
§ Hugh M.
Cartwright & Arsenij Leontjev, Use of a Genetic Algorithm
- Neural Network Hybrid algorithm in the search for high
efficiency solid-state phosphors. WSEAS Trans. on
Computers, 10, (2011), 396-406.
§ Silvia
Curteanu & Hugh M. Cartwright, Neural networks
applied in chemistry I. Determination of the optimal topology
of multilayer perceptron neural networks. J. Chemometrics,
25, (2011), 527-549. doi:10.1002/cem.1401
§ Alexander
C. Priest, Alexander J. Williamson & Hugh M. Cartwright,
The Applications of Artificial Neural Networks in the
Identification of Quantitative Structure-Activity
Relationships for Chemotherapeutic Drug Carcinogenicity. In P.
R. Cohen, N. M. Adams, & M. R. Berthold (Eds.), Advances
in Intelligent Data Analysis IX Proceedings, 6065
(2010), 137-146.
§ Hugh M.
Cartwright, Artificial neural networks in Biology and
Chemistry - the evolution of a new analytical tool, in Neural
Networks: Methods and Applications, Methods in Molecular
Biology, 458, (2008), 1-13; DJ Livingstone (ed),
Humana Press.
§ Xi Chen,
Les Sztandera, Hugh M. Cartwright, A Neural Network Approach
to the Prediction of the Glass Transition Temperature of
Polymers. Int. J. Intelligent Systems, 23, (2008),
22-32.
§ Hugh M.
Cartwright, Development and uses of Artificial Intelligence in
chemistry, in Reviews in Computational Chemistry,
Kenny Lipkowitz & Thomas R Cundari (eds.), Wiley-VCH,
(2007), 349-390.
§ C Xi, Les
Sztandera & Hugh M. Cartwright, Prediction of the glass
transition temperature of polymers using neural network and
multiple linear regression. DWI Reports, 130,
2006.
§ Hugh M.
Cartwright, Who, or what, is the teacher? How Artificial
Intelligence will control Lifelong learning. Proceedings
of the 3rd International Conference on Hands-on Science,
Manuel F. Costa & BV Dorrio (eds.), Braga, Portugal,
(2006), 576-580.
§ Hugh M.
Cartwright, Les Sztandera & Rohan M. Gunatillake, Genetic
algorithms and neural networks in the molecular design of
novel fibres, 2nd DEXA workshop: Philosophies and
Methodologies for knowledge discovery, deployment and
development of decision support systems, Krakow, Poland,
(2006).
§ Hugh M.
Cartwright & Christopher Jones, Neural Network Based
Assessment of Materials for Chemical Protection Gloves, Proceedings
of EUROCON 2003 (Computer as a Tool), Ljubljana,
Slovenia, Vol II, (2003), 319-323.
Home
page of Hugh Cartwright