Snap ML: Accelerated, Accurate, Efficient Machine Learning
Snap ML is a library for machine learning training and inference offering a familiar python scikit-learn API. With Snap ML one can typically train ML models 10 times faster than with scikit-learn on CPU and/or GPU, and achieve similar speed-up in model inference. Additionally, powerful models such as boosting machines can be trained that often beat XGBoost or LightGBM in generalization accuracy.
Haris Pozidis manages the Data and AI Systems group at IBM Research in Zurich, Switzerland. He received a Ph.D. degree in electrical engineering from Drexel University, Philadelphia, USA, in 1998, and was with Philips Research, Eindhoven, The Netherlands, before joining IBM. He has worked on read channel design for DVD and Blu-ray Disc at Philips, and played a key role in developing the first scanning probe-based data storage system at IBM, the “Millipede”. His current focus is on the development of Flash memory controllers for all-flash arrays, on AI-infused solutions for IT operations, and on accelerated software libraries for machine learning. Haris holds over 130 US patents, has co-authored more than 120 publications, is an IBM Principal Research Scientist, an IBM Master Inventor, and a Senior Member of the IEEE.