[bull-ia] Call for chapter: Optim for ML and ML for Optim [updated]

—————————————————————————————————————————————————

Call for book chapters, in:

Optimization for Machine Learning —and— Machine Learning for Optimization

 

An Open Access book.

Chapter submission deadline: May, 30th, 2018

Address for submission: https://easychair.org/conferences/?conf=oml2018

—————————————————————————————————————————————————

 

Dear colleagues,

 

Machine learning is revolutionizing our world. It is difficult to imagine another information technology that has progressed so swiftly in recent years, in terms of real impact.

 

The fields of machine learning and optimization are highly interwoven. Optimization problems forms the core of machine learning methods and modern optimization algorithms are increasingly using machine learning to improve their efficiency. This book examines the interplay between those two fields, highlighting their core similarities and how they may differ from each other on common mathematical problems.

Machine learning finds its applications in all areas of science. There are many learning methods, each of which uses a different algorithmic structure to optimize predictions based on the received data.

Hence, the first objective of this book will be to shed a light on key principles and methods that are common to both fields.

 

Machine learning and optimization share three components: representation, evaluation and iterative search. But while the optimization solvers are generally designed to be fast and accurate on implicit models, machine learning methods need to be generic and trained offline on data sets. Machine learning problems present new challenges to the optimization researchers, and machine learning practitioners seek simpler, generic optimization algorithms.

This book is thus focused on a common field of research: how to solve new machine learning problems with robust optimization solvers and how to use new optimization methods for existing machine learning problems.

 

Quite recently, modern approaches to machine learning have also been applied to the design of optimization algorithms themselves, taking advantage of their ability to capture valuable information from complex structures in large spaces. Those capacities appear to be useful, especially for the representation and evaluation components. As large complex structures are ubiquitous in optimization problems and can be used as huge implicit data sets, the use of machine learning allowed improvements in efficiency and genericity of optimization solvers.

This book aims at introducing modern advances in algorithm selection, configuration and engineering that rely on machine learning.

 

If you conduct interesting research in those domains and want to share them, or if you are willing to share your global view, do not hesitate to contribute to this open access initiative!

Chapters can be written in french and/or in english and will be translated.

See attached files for more details and for the call in french.

 

Important dates

– Chapter submission:                    May, 30th, 2018

– Notification to authors:              June, 30th, 2018

– Camera-ready papers due:         July, 30th, 2018

– Publication:                                    November, 15th, 2018

 

The book will be published by Wiley (translated in english) and ISTE (translated in french).

 

Contacts for more information

rachid.chelouah@eisti.eu

johann.dreo@thalesgroup.com

 

Sincerely,

Dr. Johann Dreo

Ph.D. Artificial Intelligence Algorithmics

Thales, Research & Technology

 

[@@ THALES GROUP INTERNAL @@]