skip to content

Laboratory for Scientific Computing

 

Biography

Anita Faul came to Cambridge after studying two years in Germany. She did Part II and Part III Mathematics at Churchill College, Cambridge. Since these are only two years, and three years are necessary for a first degree, she does not hold one. However, this was followed by a PhD on the Faul-Powell Algorithm for Radial Basis Function Interpolation under the supervision of Professor Mike Powell. This collaboration resulted in an Erdős number of 4 (Powell - Beatson - Chui - Erdős). After this she worked on the Relevance Vector Machine with Mike Tipping at Microsoft Research Cambridge. Ten years in industry followed where she worked on various algorithms on mobile phone networks, image processing and data visualization. Current projects are on machine learning techniques. In teaching she enjoys to bring out the underlying, connecting principles of algorithms which is the emphasis of her book on Numerical Analysis. She is working on a book on machine learning.

ORCID iD iconorcid.org/0000-0002-5911-2109

https://www.researchgate.net/profile/Anita_Faul

Research Interests

There are several challenges with which data present us nowadays. For one there is the abundance of data and the necessity to extract the essential information from it. When tackling this task a balance has to be struck between putting aside irrelevant information and keeping the relevant one without getting lost in detail, known as over-fitting. The law of parsimony, also known as Occam’s razor should be a guiding principle, keeping models simple while explaining the data.

The next challenge is the fact that the data samples are not static. New samples arrive constantly through the pipeline. Therefore, there is a need for models which update themselves as the new sample becomes available. The models should be flexible enough to become more complex should this be necessary. In addition the models should inform us which samples need to be collected so that the collection process becomes most informative.

Another challenge are the conclusions we draw from the data. After all, as popularized by Mark Twain: "There are three kinds of lies: lies, damned lies, and statistics." An objective measure of confidence is needed to make generalized statements.

The last challenge is the analysis. Can we build systems which inform us of the underlying structure and processes which gave rise to the data? Moreover, it is not enough to discover the structure and processes, we also need to add meaning to it. Here different disciplines need to work together.

-->

Moodle

Fundamentals, Interpolation and Non-Linear Systems
Linear Systems
Numerical Integration and ODEs
Numerical Differentiation and PDEs
Machine Learning
Machine Learners

Publications

Key publications: 

A Concise Introduction to Machine Learning Machine Learning is known by many different names, and is used in many areas of science. It is also used for a variety of applications, including spam filtering, optical character recognition, search engines, computer vision, NLP, advertising, fraud detection, robotics, data prediction, astronomy. Considering this, it can often be difficult to find a solution to a problem in the literature, simply because different words and phrases are used for the same concept. This class-tested textbook aims to alleviate this, using mathematics as the common language. It covers a variety of machine learning concepts from basic principles, and illustrates every concept using examples in MATLAB For further information, visit the publishers' webpage.

A Concise Introduction to Numerical AnalysisThis textbook provides an accessible and concise introduction to numerical analysis for upper undergraduate and beginning graduate students from various backgrounds. It was developed from the lecture notes of four successful courses on numerical analysis taught within the MPhil of Scientific Computing at the University of Cambridge. The book is easily accessible, even to those with limited knowledge of mathematics.

Students will get a concise, but thorough introduction to numerical analysis. In addition the algorithmic principles are emphasized to encourage a deeper understanding of why an algorithm is suitable, and sometimes unsuitable, for a particular problem.

Additional material such as the solutions to odd numbered exercises and MATLAB® examples can be downloaded from the publishers' webpage. The book can be purchased here.

  • "Semi-supervised Learning with Graphs: Covariance Based Superpixels for Hyperspectral Image Classification". P. Sellars, A. Aviles-Rivero, N. Papadakis, D. Coomes, A. Faul, C.-B. Schönlieb
  • "Deep Learning Applied to Seismic Data Interpolation". A. Mikhailiuk, A Faul, European Association of Geoscientists and Engineers (EAGE), IEEE (2018).
  • ''Bayesian Feature Learning for Seismic Compressive Sensing and Denoising", G. Pilikos, A.C. Faul, Geophysic (2017).
  • "Seismic compressive sensing beyond aliasing using Bayesian feature learning", G. Pilikos, A.C. Faul and N. Philip, 87th Annual International Meeting, SEG, Expanded Abstracts (2017).
  • "Relevance Vector Machines with Uncertainty Measure for Seismic Bayesian Compressive Sensing and Survey Design", G. Pilikos, A.C. Faul , IEEE International Conference on Machine Learning and Applications (2016).
  • "The model is simple, until proven otherwise - how to cope in an ever changing world", A.C. Faul, G. Pilikos, Data for Policy (2016).
  • "A Krylov subspace algorithm for multiquadric interpolation in many dimensions", A.C. Faul, G. Goodsell and M.J.D. Powell, published in IMA Journal of Numerical Analysis (2005).
  • "Fast marginal likelihood maximisation for sparse Baysian models", M.E. Tipping, A.C. Faul, published in Proceedings of the Ninth International= Workshop on Artificial Intelligence and Statistics (2003).
  • "Analysis of Sparse Bayesian Learning", A.C. Faul, M.E. Tipping, published in Advances in Neural Information Processing Systems 14 (2002).
  • "A variational approach to robust regression", A.C. Faul, M.E. Tipping, published in the Proceedings of ICANN'01.
  • "Proof of convergence of an iterative technique for thin plate spline interpolation in two dimensions", A.C. Faul, M.J.D. Powell, published in Advances in Computational Mathematics, Vol. 11.
  • "Krylov subspace methods for radial basis function interpolation", A.C. Faul, M.J.D. Powell, published in Numerical Analysis, (1999).
  • "Iterative techniques for radial basis function interpolation", Ph.D. thesis.
Teaching Associate
Fellow, Director of Studies in Mathematics and Graduate Tutor at Selwyn College

Contact Details

+44 (0)1223 337273
Not available for consultancy

Affiliations

Classifications: