Bergen-ML meetup 2018.10.31

17.00 - We meet for some food and mingling

17.30 - Rogério C. B. L. Póvoa has a Phd in Computural Intelligence – optimization and machinelearning from PUC – Rio. He has experience as a researcher and consultant from Brasil. He has now moved to Norway where he is consultant for Apparat. Rogério will present his Ph.D. work: "Development of unimodal and multimodal optimization algorithms based on multi-gene genetic programming". He has developed some numerical optimization algorithms using genetic programming and applied one of them to optimize a quantum well infrared photodetector. The results presented new structures better than those known in the literature with an improvement of 59.09%.

18.00 - Break

18.15 - Berent Å. S. Lunde is a current PhD student at UiS with a master’s degree in statistics from UiB, who has previously been working as an actuary (Actuary of the year 2017!).
Berent will present his work on gradient tree boosting; a set of algorithms that have dominated a large part of machine learning competitions the last five years. "By coupling information theory, the frequency domain and tree-boosting, the algorithm can adaptively learn the size of the tree, and how many trees that should be added. This particular characteristic is nice because of the model’s robustness, the computational power needed to train the algorithm and the facilitation of democratization of machine learning"

18.45 Q&A

19.00 - Thank you and see you next time