MIP Modelling Made Manageable

Event details

  • When: 19th June 2019 11:00 - 12:00
  • Where: Cole 1.33a
  • Series: AI Seminar Series
  • Format: Lecture, Seminar

Can a user write a good MIP model without understanding linearization? Modelling languages such as AMPL and AIMMS are being extended to support more features, with the goal of making MIP modelling easier. A big step is the incorporation of predicates, such a “cycle” which encapsulate MIP sub-models. This talk explores the impact of such predicates in the MiniZinc modelling language when it is used as a MIP front-end. It reports on the performance of the resulting models, and the features of MiniZinc that make this possible.

Professor Mark Wallace is Professor of Data Science & AI at Monash University, Australia. We gratefully acknowledge support from a SICSA Distinguished Visiting Fellowship which helped finance his visit.

Professor Wallace graduated from Oxford University in Mathematics and Philosophy. He worked for the UK computer company ICL for 21 years while completing a Masters degree in Artificial Intelligence at the University of London and a PhD sponsored by ICL at Southampton University. For his PhD, Professor Wallace designed a natural language processing system which ICL turned into a product. He moved to Imperial College in 2002, taking a Chair at Monash University in 2004.

His research interests span different techniques and algorithms for optimisation and their integration and application to solving complex resource planning and scheduling problems. He was a co-founder of the hybrid algorithms research area and is a leader in the research areas of Constraint Programming (CP) and hybrid techniques (CPAIOR). The outcomes of his research in these areas include practical applications in transport optimisation.

He is passionate about modelling and optimisation and the benefits they bring.  His focus both in industry and University has been on application-driven research and development, where industry funding is essential both to ensure research impact and to support sufficient research effort to build software systems that are robust enough for application developers to use.

He led the team that developed the ECLiPSe constraint programming platform, which was bought by Cisco Systems in 2004. Moving to Australia, he worked on a novel hybrid optimisation software platform called G12, and founded the company Opturion to commercialise it.  He also established the Monash-CTI Centre for optimisation in travel, transport and logistics.   He has developed solutions for major companies such as BA, RAC, CFA, and Qantas.  He is currently involved in the Alertness CRC, plant design for Woodside planning, optimisation for Melbourne Water, and work allocation for the Alfred hospital.

Professor Aaron Quigley new SICSA Director

Congratulations to Professor Aaron Quigley who has been appointed as the new Director of SICSA. Aaron, the Chair of Human Computer Interaction co-founded SACHI, the St Andrews Computer Human Interaction research group and served as its director from 2011-2018.

In his volunteer roles he is the ACM SIGCHI Vice President for Conferences (on the ACM SIGCHI Executive Committee), member of the ACM Europe Council Conferences Working Group, a board member of ScotlandIS and an ACM Distinguished Speaker. Aaron will be general co-chair for the ACM CHI conference in Asia in 2021.

For more information about Professor Quigley, please see https://aaronquigley.org.

Graduation Reception: Wednesday 26th June

Event details

  • When: 26th June 2019 11:00 - 13:00
  • Where: Cole Coffee Area
  • Format: graduation

The School of Computer Science will host a graduation reception on Wednesday 26th June, in the Jack Cole building, between 11.00 and 13.00. Graduating students and their guests are invited to the School to celebrate with a glass of bubbly and a cream cake. Computer Science degrees will be conferred in an afternoon ceremony in the Younger Hall. Family and friends who can’t make it on the day can watch a live broadcast of graduation. Graduation receptions have been held in the school from 2010.

A class photo will be taken at 12.00 outside the Jack Cole building.

Juho Rousu: Predicting Drug Interactions with Kernel Methods

Event details

  • When: 30th April 2019 14:00 - 15:00
  • Where: Cole 1.33a
  • Format: Seminar

Title:
Predicting Drug Interactions with Kernel Methods

Abstract:
Many real world prediction problems can be formulated as pairwise learning problems, in which one is interested in making predictions for pairs of objects, e.g. drugs and their targets. Kernel-based approaches have emerged as powerful tools for solving problems of that kind, and especially multiple kernel learning (MKL) offers promising benefits as it enables integrating various types of complex biomedical information sources in the form of kernels, along with learning their importance for the prediction task. However, the immense size of pairwise kernel spaces remains a major bottleneck, making the existing MKL algorithms computationally infeasible even for small number of input pairs. We introduce pairwiseMKL, the first method for time- and memory-efficient learning with multiple pairwise kernels. pairwiseMKL first determines the mixture weights of the input pairwise kernels, and then learns the pairwise prediction function. Both steps are performed efficiently without explicit computation of the massive pairwise matrices, therefore making the method applicable to solving large pairwise learning problems. We demonstrate the performance of pairwiseMKL in two related tasks of quantitative drug bioactivity prediction using up to 167 995 bioactivity measurements and 3120 pairwise kernels: (i) prediction of anticancer efficacy of drug compounds across a large panel of cancer cell lines; and (ii) prediction of target profiles of anticancer compounds across their kinome-wide target spaces. We show that pairwiseMKL provides accurate predictions using sparse solutions in terms of selected kernels, and therefore it automatically identifies also data sources relevant for the prediction problem.

References:
Anna Cichonska, Tapio Pahikkala, Sandor Szedmak, Heli Julkunen, Antti Airola, Markus Heinonen, Tero Aittokallio, Juho Rousu; Learning with multiple pairwise kernels for drug bioactivity prediction, Bioinformatics, Volume 34, Issue 13, 1 July 2018, Pages i509–i518, https://doi.org/10.1093/bioinformatics/bty277

Short Bio:
Juho Rousu is a Professor of Computer Science at Aalto University, Finland. Rousu obtained his PhD in 2001 form University of Helsinki, while working at VTT Technical Centre of Finland. In 2003-2005 he was a Marie Curie Fellow at Royal Holloway University of London. In 2005-2011 he held Lecturer and Professor positions at University of Helsinki, before moving to Aalto University in 2012 where he leads a research group on Kernel Methods, Pattern Analysis and Computational Metabolomics (KEPACO). Rousu’s main research interest is in learning with multiple and structured targets, multiple views and ensembles, with methodological emphasis in regularised learning, kernels and sparsity, as well as efficient convex/non-convex optimisation methods. His applications of interest include metabolomics, biomedicine, pharmacology and synthetic biology.

PhD viva success: Evan Brown

Congratulations to Evan Brown, who successfully defended his thesis today. He is pictured with Internal examiner Dr Tristan Henderson and external examiner Professor Chris Marsden, Professor of Internet Law at the University of Sussex.

Evan’s PhD research on using corpus linguistics to build collaborative legal research tools was supervised by Professor Aaron Quigley.

Continued success for MSc student Jessica Cooper

The work of our MSc student, Jessica Cooper, supervised by Oggie Arandjelovic on the use of deep learning for the analysis of ancient Roman coins has been attracting widespread attention. From tech media to web sites of history, heritage, and numismatics focused communities, Jessica’s work has been recognized as highly innovative, with a potential to change the direction of research in the area. Jessica will be rejoining St Andrews in a month’s time, working with Oggie Arandjelovic on deep learning in pathology image analysis.

Best paper finalist award for Xingzhi Yue and Neofytos Dimitriou

A paper describing the work of our MSc student Xingzhi Yue and PhD student Neofytos Dimitriou, supervised by Oggie Arandjelovic and in collaboration with the School of Medicine, gets the best paper finalist award at the latest International Conference on Bioinformatics and Computational Biology (BICOB 2019). The key contribution of the work is a novel deep learning based algorithm for the analysis of extremely large pathology image slides, capable of automating and improving colorectal cancer prognosis.

Distinguished Lecture Series: Formal Approaches to Quantitative Evaluation

Event details

  • When: 8th April 2019 09:30 - 15:30
  • Where: Lower College Hall
  • Series: Distinguished Lectures Series
  • Format: Distinguished lecture

Biography:
Jane Hillston was appointed Professor of Quantitative Modelling in the School of Informatics at the University of Edinburgh in 2006, having joined the University as a Lecturer in Computer Science in 1995. She is currently Head of the School of Informatics. She is a Fellow of the Royal Society of Edinburgh and Member of Academia Europaea. She currently chairs the Executive Committee of the UK Computing Research Committee.
Jane Hillston’s research is concerned with formal approaches to modelling dynamic behaviour, particularly the use of stochastic process algebras for performance modelling and stochastic verification. The application of her modelling techniques have ranged from computer systems, to biological processes and transport systems. Her PhD dissertation was awarded the BCS/CPHC Distinguished Dissertation award in 1995 and she was the first recipient of the Roger Needham Award in 2005. She has published over 100 journal and conference papers and held several Research Council and European Commission grants.
She has a strong interest in promoting equality and diversity within Computer Science; she is a member of the Women’s Committee of the BCS Computing Academy and chaired the Women in Informatics Research and Education working group of Informatics Europe 2016—2018, and during that time instigated the Minerva Informatics Equality Award.

Formal Approaches to Quantitative Evaluation
Qualitative evaluation of computer systems seeks to ensure that the system does not exhibit bad behaviour and is in some sense “correct”. Whilst this is important it is also often useful to be able to reason not just about what will happen in the system, but also the dynamics of that behaviour: how long it will take, what are the probabilities of alternative outcomes, how much resource is used….? Such questions can be answered by quantitative analysis when information about timing and probability are incorporated into models of system behaviour.

In this short series of lectures I will talk about how we can extend formal methods to support quantitative evaluation as well as qualitative evaluation of systems. The first lecture will focus on computer systems and a basic approach based on the stochastic process algebra PEPA. In the second lecture I will introduce the language CARMA which is designed to support the analysis of collective adaptive systems, in which the structure of the system may change over time. In the third lecture I will consider systems where the exact details of behaviour may not be known and present the process algebra ProPPA which combines aspect of machine learning and inference with formal quantitative models.

Timetable:
Lecture 1: 9:30 – 10:30 – Performance Evaluation Process Algebra (PEPA)

Coffee break at 10:30 – 11:15
Lecture 2: 11:15 – 12:15 – Collective Adaptive Resource-sharing Markovian Agents (CARMA)

Lecture 3: 14:15 – 15:15 – Probabilistic Programming for Stochastic Dynamical Systems (ProPPA)


Venue: Upper and Lower College Halls

Encoding Egyptian quadrats in Unicode

Unicode 12, released 5th March 2019, includes 9 control characters for Ancient Egyptian hieroglyphic text. These resulted from an initiative by Dr. Mark-Jan Nederhof (St Andrews) and Egyptologists at the University of Liège, CNAM (Paris) and the Berlin-Brandenburgische Akademie der Wissenschaften, in collaboration with Unicode experts. The control characters allow hieroglyphs to be arranged horizontally and vertically much as in original inscriptions. This removes the foremost obstacle to adoption of Unicode in Egyptology.

The control characters:
https://www.unicode.org/charts/PDF/Unicode-12.0/U120-13430.pdf

Although existing fonts are not yet able to interpret the control characters directly, hieroglyphic text can now be displayed on web pages with the help of JavaScript:
https://mjn.host.cs.st-andrews.ac.uk/egyptian/res/js/

Lao Characters for Pali added to Unicode 12

Congratulations to Vinodh Rajan, Ben Mitchell, Martin Jansche and Sascha Brawer on their successful proposal for additions to the repertoire of ISO/IEC 10646, which will see Pali letters added to Lao in Unicode 12. As a result, it is now possible to write both Pali/Sanskrit in Lao and represent the entire Tripitaka in the Lao script. The proposal (https://bit.ly/2TE2XKJ) submitted in 2017 was finally added to the Unicode standard this year.

Vinodh explained that the proposal allows four things. Firstly, one can now transcribe liturgical Pali (the liturgical language of Theravada Buddhism) texts and by extension the whole Pali Tripitaka (the Theravada Buddhist canon) in the Lao script without any distortion, providing lay people accurate access to these liturgical texts. Previously, the texts had to go through some sort of distortion due to the lack of appropriate characters, which means they had to be approximated. Secondly, it allows people who would want to use etymological orthography for Lao (it currently uses a phonemic orthography) access to the necessary additional characters. Thirdly, there are several books printed (mostly in the 1930’s) using the expanded alphabet that need to be eventually digitized. This will enable their proper digitization by allow plain-text representation of all the Lao characters. Lastly, it will improve the transliteration accuracy between Lao and neighboring scripts like Thai and Khmer.

The expanded Lao alphabet can be found here:
http://aksharamukha.appspot.com/#/describe/LaoPali

Vinodh, a St Andrews Computer Science alumnus completed his PhD in 2016. His thesis, Quantifying scribal behavior : a novel approach to digital paleography was supervised by Dr Mark-Jan Nederhof.