This is the first announcement of the Summer School on Machine Learning for High Energy Physics 2016, to be held at Lund, Sweden, June 20-26 2016 as a satellite event of LHC Physics Conference.
Early registration deadline is 30 April, 2016.
The primary goal of MLHEP school will be a focused introduction to modern machine learning techniques that could improve physics performance for variety of HEP-related problems. School pays attention to student experience, so along with “hands-on” seminars a dedicated data science competition will be organized.
Additionally, the school will include series of talks that show real examples of improvements for particular physics cases due to machine learning techniques. It is ideally suited for advanced graduate students and young postdocs willing to learn how to
- formulate HEP-related problem in ML-friendly terms,
- select quality criteria for given problem,
- understand and apply principles of widely-used classification models (e.g. boosting, bagging, BDT, neural networks, etc) to practical cases,
- optimize features and parameters of given model in efficient way under given restrictions,
- select the best classifier implementation amongst variety of ML libraries (scikit-learn, xgboost, deep learning libraries, etc), and
- define and conduct reproducible data-driven experiments.