A study to be presented at the IEEE Congress on Evolutionary Computation 2019 in Wellington, New Zealand shows how the performance potential of elite athletes can be improved five-fold using an approach to Artificial Intelligence inspired by biological evolution. The research is a collaboration between STATSports and the UCD Natural Computing Research & Applications Group as part of the Science Foundation Ireland funded Lero Software Research Centre undertaken by Mark Connor, Dr David Fagan and Professor Michael O’Neill.
Today we are sad to see one of the original founding members of the NCRA, Dr Róisín Loughran depart to take up a new opportunity in Dundalk Institute of Technology. On the other hand we are delighted that she joins our Lero colleagues and collaborators in the Regulated Software Research Centre and we look forward to working with her in the future.
UCD Natural Computing Research & Applications Group researcher Nam Le has been listed as an EvoStar Outstanding Student of 2019 and is shortlisted for the EvoStar 2019 Best Student Paper Award for his research on “The Evolution of Self-taught Neural Networks in a Multi-agent Environment“. Nam is investigating the interplay between Learning and Evolution as means of adaptation in complex environments.
Professor Michael O’Neill and Dr David Fagan have published “The Elephant in the Room – Towards the Application of Genetic Programming to Automatic Programming” where they discuss how the field of Genetic Programming , representing one of the largest literatures in Artificial Intelligence, should go back to its roots and become re-focused towards the problem of automatic software creation. The research was presented at the GPTP Workshop in Ann Arbor, Michigan in May 2018 and has recently been published in a book by Springer (Genetic Programming Theory and Pratice XVI) in 2019.
Researchers from the UCD Natural Computing Research & Applications Group in collaboration with Bell Labs published in the IEEE/ACM Transactions on Networking journal (Vol.27 Issue 1) a paper titled “Automated Self-Optimization in Heterogeneous Wireless Communications Networks” where they demonstrate that their approach to Artificial Intelligence based on Evolutionary Computation more than triples peak data rates, and show how their approach can be used by network operators to flexibly tune the tradeoff between peak rates and fairness across all mobile network users. The research is supported by Science Foundation Ireland through an Investigator Award to Prof Michael O’Neill who is a Full Professor of Business Analytics at the UCD School of Business.
12-14 December we welcomed TPNC 2018, the 7th instance of the international conference on the Theory & Practice of Natural Computing with over 40 delegates to the UCD Natural Computing Research & Applications Group in the UCD Michael Smurfit Graduate Business School. The proceedings are published by Springer in Lecture Notes in Computer Science Volume LNCS 11324.
Congratulations to David Lynch and Stefan Forstenlechner who both successfully defended their PhD theses this month on “Automated Self-Optimisation in Heterogeneous Wireless Communications Networks using Techniques from Evolutionary Computation” (Lynch) and “Program Synthesis with Grammars and Semantics in Genetic Programming” (Forstenlechner).
John Woodward (Queen Mary University London) visited our group on 9 November 2018 to give a seminar on “Genetic Improvement: Taking Real-World Source Code and Improving it Using Genetic Programming”
Prof Michael O’Neill in collaboration with the Analytics Institute and supported by EY launched the inaugural National Business Analytics Maturity Study at the National Analytics Conference held in the Mansion House Dublin on 6 November 2018
Welcome to Tomasz Pawlak who is a visiting researcher with the NCRA group funded by a grant from the National Science Centre, Poland, grant no. 2016/23/D/ST6/03735. Tomasz will spend two weeks at UCD working with the NCRA group before returning to Poznan University of Technology in Poland. His research interests are synthesis of mathematical programming models from data, program synthesis using genetic programming, and process mining.