Machine Learning for Geoscientists with Hands-on Coding
By: Ehsan Naeini
Instructor
Dr. Ehsan Naeini
(Earth Science Analytics,
United Kingdom)
Duration
7–8 July 2020:
4:00PM-8:00PM CEST
4 hours/day
Disciplines
Data Science - Machine Learning
Course format
The EAGE Interactive online short courses bring carefully selected courses of experienced instructors from industry and academia online to give participants the possibility to follow the latest education in geoscience and engineering remotely. The courses are designed to be easily digested over the course of two or three days. Participants will have the possibility to interact live with the instructor and ask questions.
Course description
Machine learning has been around for decades or, depending on your view, centuries. By applying machine learning to our workflows, e.g. petrophysics, rock physics, seismic processing and reservoir characterization, we can bring speed, efficiency and consistency over traditional methods of data analysis. In addition, we can implement a range of machine learning techniques together with optimization algorithms and statistics to identify new patterns and relationships in multi-dimensional datasets. This has the potential to enhance our quantification and strengthen our interpretation of the subsurface; ultimately leading to a more accurate predictive outcome.
In this course we attempt to layout the reality of artificial intelligence, machine learning, deep learning and big data. We cover the basic principles of machine learning and some of the most widely used algorithms. We continue by explaining a workflow for implementing a typical machine learning application in practice and to quality control and interpret the outcomes. Following this we shift focus to Geoscience and show various examples in which machine learning algorithms have been implemented for well- and/or seismic-based applications. Given the hands-on coding nature of this course, trainees will code up a classification and a regression algorithm for lithology/facies and well log prediction correspondingly. Throughout these exercises the trainees will become familiar with the flexibility of coding machine learning in Python (although we do not intend to teach Python in details in this course) as well as familiarization with publicly available python libraries for machine learning and analytics. The course is for entry level practitioners and involves hands-on coding, hence having some Python skills is an advantage but not essential.
Course objectives
- Use python
- Understand various machine learning algorithms, concepts and terminologies
- Capable of analyzing data in big scales
- QC for machine learning applications
- Extend their newly learned knowledge to their day to day practice and implement their own ideas
Course outline
- Introduction
- Machine Learning Principles
- Machine Learning in Practice
- Exercise 1: ML for classification
- Exercise 2: ML for regression
Participants' profile
The course is designed for basically everyone.
Prerequisites
There are no prerequisites, but basic Python knowledge can be useful.
About the instructor
Ehsan Naeini
Dr. Ehsan Naeini is a Geoscience researcher and practitioner with more than 13 years' industry experience, particularly in seismic inversion, processing, computational and data science. He has an MSc and PhD in Geophysics (Exploration Seismology) from the University of Tehran and a BSc in Physics from the University of Isfahan. Whilst studying for his PhD, Ehsan was a lecturer in Geophysics at the University of Isfahan.
Ehsan is the Chief Product Officer at Earth Science Analytics and is leading the development of EarthNET, a petroleum geoscience platform based and AI, machine learning and cloud technology while working at the intersection of sales, marketing, client support and service project execution. Prior to joining Earth Science Analytics, Ehsan was VP Research & Development at Ikon Science. He has also been research geophysicist and senior researcher at CGG. He has taught ML courses to various groupings sponsored by AAPG, SEG, Royal Geological Society and at Mines.