Choose timezone
Your profile timezone:
Active Learning in JMP
→
Europe/Amsterdam
Description
Active Learning in JMP
Part of the ENBIS-24 Leuven conference.
Instructors
Chris Gotwalt & Phil Kay, JMP
Outline
In industrial R&D and process improvement we experiment to understand wickedly complex, multifactor systems. Empirical models of our processes enable us to bring new technologies to market and deliver consistent quality for our customers. Statistical Design and Analysis of Experiments (DOE) is a proven methodology for efficiently capturing the data to build the required model of your process or system. More recently there has been interest in applying "Bayesian Optimization" or "Active Learning" in process and formulation development. These sequential experimentation methods promise greater speed and a simpler workflow for non-statisticians by prioritising goal-seeking over model-building.
JMP, as the leading DOE software and a complete data science tool for scientists and engineers, is the ideal environment for exploring these different approaches. This hands-on workshop with JMP Chief Data Scientist, Chris Gotwalt, will be a unique opportunity to see case studies on "model-agnostic" experimentation. All participants will get free access to pre-install JMP and a special Active Learning addin. Attendees are invited to bring their Windows or Mac computers to try out capabilities including Space-Filling Designs, Design Augmentation, JMP's Prediction Profiler and Gaussian Process Models. Pre-knowledge in statistics, DOE or JMP are not required.
Short bios
Chris Gotwalt leads the statistical software development and testing teams for JMP Statistical Discovery. His passion is developing new technologies that accelerate innovation in industry and science. Since joining the company as a PhD student intern in 2001, Gotwalt has contributed many numerical algorithms and new statistical techniques. He has authored algorithms in JMP for fitting neural networks, linear mixed models, optimal design of experiments, analytical procedures for text analysis, and the algorithms for fitting structural equation models. Gotwalt is a principal investigator for Self-Validating Ensemble Models (SVEM), a procedure that makes machine learning possible for the small data sets often encountered in industry. He holds adjunct professorial positions at North Carolina State University, University of Nebraska and University of New Hampshire, and was the 2020 Chair of the Quality and Productivity Section of the American Statistical Association.
Phil Kay leads JMP's Global Technical Enablement Team and he loves showing people how data analytics enables better science. His passion for this started with some big successes using statistical design of experiments as a development chemist at FujiFilm. Phil has a master’s degree in applied statistics and a master’s and PhD in chemistry. He is a chartered chemist and chair of the Royal Sociey of Chemistry Process Chemistry and Technology Interest group.
Registration
Course -- Registration