Login
Create Account
Offices
Tulsa
Washington DC
London
Dubai
Singapore
Bogotå
Lagos
Divisions
DPA
EMD
DEG
PSGD
Foundation
Resources
GIS Open Files
Videos
Gateways
Podcast
Sitemap
Join, Renew or Reinstate
membership
Vote Now
election
Menu
Home
Member Focus
YoungPros Focus
Students Focus
Women's Network
Environmental Focus
Economics Focus
About
AAPG
Overview
Annual Reports
Committees
Honors and Awards
Programs
Affiliated Societies
Leadership
Executive Committee
Officer Candidates
Advisory Council
House of Delegates (HoD)
HoD Honors & Awards
Sections
Regions
Divisions
DPA
EMD
DEG
PSGD
DPA Directory
Membership
Types of Membership
Member
Associate
Student
Benefits and Services
Comparison Chart
Prospective Members
Graduated Dues
Recruitment
Constitution and Bylaws
Code of Ethics
Member Directory
Transfer to Member
Petroleum Geology
Petroleum through Time
What Is Petroleum?
Petroleum Exploration History
Museums and Historical Sites
World of Petroleum
Oil and Gas Prices
Careers in the Petroleum Industry
Exploration Today
U.S. Offshore Oil & Gas Resources
Petroleum Technology
Petroleum Products
Tools of the Trade
Refining and Chemicals
Petroleum Chemistry
Anatomy of an Oil Accumulation
Geology and Petroleum
Sedimentology and Stratigraphy
Structural Geology
Paleontology and Biostratigraphy
Geochemistry
Seismology and Geophysics
Plate Tectonics
Paleogeography
Geologic Maps
Science
Discipline
Structure
Compressional Systems
Geomechanics and Fracture Analysis
Extensional Systems
Fold and Thrust Belts
Salt Tectonics
Structural Analysis - Other
Tectonics (General)
Basin Modeling and Geochemistry
Basin Modeling
Maturation
Migration
Oil and Gas Analysis
Oil Seeps
Petroleum Systems
Source Rock
Thermal History
Engineering
Drilling
Completion and Production
Reservoir Engineering
Geophysics
Direct Hydrocarbon Indicators
Gravity
Magnetic
Seismic
Sedimentology and Stratigraphy
Carbonates
Clastics
Evaporites
Sequence Stratigraphy
Business and Economics
Economics
Resource Estimation
Risk Analysis
Environmental
Petrophysics and Well Logs
Astrogeology
Play Types
Coalbed Methane
Deep Basin Gas
Diagenetic Traps
Fractured Carbonate Reservoirs
Bitumen/Heavy Oil (Oil Sands)
Oil Shale
Shale Gas
Stratigraphic Traps
Structural Traps
Subsalt Traps
Tight Gas Sands
Alternative Resources
Coal
Gas Hydrates
Geothermal
Renewable Energy
Uranium (Nuclear)
Geographic
Countries
Interests
Arctic
Antarctica
Drones
Machine Learning
Petroleum Resources Management System
CO2 Storage Resources Management System
Career
Training
In Person
Field Seminars
Forums
Short Courses
Visiting Geoscientists
Workshops
Online
Distinguished Lecture Program
Certificate Programs
Certificate Courses
e-Symposia
Ethics Videos (DPA)
Traditional Courses
Virtual Events
Past Virtual Events
Instructors
DPA PDH Form
Jobs
Career Center
Career Resources
Classifieds
Counselors and Recruiters
Job Boards
Member Registry
Expert Witness (DPA)
Consulting Expertise
AAPG N.E.T.
Technical Interest Groups (TIGs)
Special Interest Groups (SIGs)
All Discussion Boards
Discussion Boards FAQ
Participate
Propose Training
Join a Committee
Present a Paper
Imperial Barrel Award
AAPG Women's Network
AAPG Award Nominations
EC Officer Nominations
Events
Conferences
International Meeting for Applied Geoscience & Energy
International Conference & Exhibition
Energy Opportunities Conference
Carbon Capture, Untilization and Storage
URTeC
OTC
IPTC
MEOS-GEO
Expos
BEOS
NAPE
Summer NAPE
Training
Field Seminars
Short Courses
Workshops
Research
Hedberg Conferences
Research Symposia
Event Listings
Calendar
Virtual Events
Exhibit Opportunities
Sponsorship Opportunities
Publications
Journals
Bulletin
Current Issue
Archives
Search
Interpretation
Environmental Geosciences
Current Issue
Archives
News
Explorer
Current Issue
Archives
Classifieds
Newsletters
Correlator
Current Issue
Archives
Delegates Voice
Current Issue
Archives
Special Publications
Books
Archie
Hedberg
Memoirs
Methods in Exploration
Special Publications
Studies in Geology
Book Reviews
CD-DVD
Maps
Digital Downloads
AAPG Wiki
Online
Datapages
Datapages Archives
Collections
Search and Discovery
GIS Open Files
GIS Map Publishing Program
Blogs
Events Blog
Learn! Blog
wwwUpdate
Foundation Update
Advertise
Media
Copyright
Submit
Proposal
Instructions
Copyright Information
Store
Help
Home
Career
AAPG N.E.T.
Group Activity
Login to Join
Deep Learning/Machine Learning Technical Interest Group (TIG)
Applying new analytics, neural networks, computational approaches using structured and unstructured data, and also training neural networks with supervised and unsupervised algorithms. Chaired by Patrick Ng and Andrew Munoz.
Deep Learning - Machine Learning TIG
Prev
Go to previous topic
Next
Go to next topic
Forums
Technical Interest Groups
Deep Learning - Machine Learning TIG Discussions
Last Post 02 Oct 2020 05:21 PM by
Patrick Ng
Highlights on Tackle the Issues - Where should Machine Learning go (and not go)?
0 Replies
Sort:
Oldest First
Most Recent First
You are not authorized to post a reply.
Author
Messages
Patrick Ng
Basic Member
Posts:148
02 Oct 2020 05:21 PM
Observations - Dr. Lian and speaker Ball showed in examples simple architecture, random forest (RF) and support vector machine (SVM) work well in many cases (some as good as deep learning models).
Q1: what does that tell us about ML model - simplicity vs complexity, and the direction?
Recap - much has to do with data / sampling. If we have lots of data, deep learning will do well. When we have limited data, RF and SVM may perform better (in supervised and unsupervised ML, respectively). Deep learning delivers benefits when we show tender loving care on input data. (Note in Making the Business Case for Geology on Day 2 - emphasize the key to success of bringing in cores, geology, petrophysics etc. together early on vs working on data in silos then integrating information at late stage. As such, RF deemed the best vs neural network and others.)
It echoes Prof Andrew Ng @Stanford (founder of Coursera and deeplearnnig.ai). Often the success of ML projects hinges not on size of big data (say lots and lots, bigger than Texas big), but “strong” data (multidiscipline data fit that mode). Once again, I highly recommend Coursera’s Machine Learning* by Andrew. He shares real life project insights from Silicon Valley to Silicon Alley (Beijing). For $49, it was my best investment in time and money.
*his keystone course led to the launch of Coursera and validated MOOC as a business.
Q2 how soon do you think a fully AI-interpreted prospect will be drilled (per opening Michel T Halbouty lecture)?
Kathy succinctly pointed out that autonomous drilling on prospect is underway. It is happening and only a matter of time before it become routine, once the kinks are ironed out in that process.
Sashi opined that once we have every bit of data in the cloud, including sensors, the infrastructure is there for that to happen.
My take - the choice is not binary. Duval’s closing thoughts (from Michel T Halbouty lecture) puts it nicely, “like drilling the first well in a new basin, we bring together different disciplines, adjust our thought process and navigate the new normal.”
Q3: given the half life* is 3-6 months, how do we catch up on implementing ML in our projects, like we run and try to catch the ball while learning how to do it?
Panel consensus is that we shall think of life-long learning (project life) and be diligent in evaluating data, ML workflow and outcomes. As Renato talked about DNA / human-centered AI. To make progress, we shall embrace the risk of failure (ML may not work in all situations) and learn from it. Make changes to the model, and how we prepare data (one in audience suggested ML for data QC). Alec chimed in "not do the same thing over and over, and expect a different outcome. That will be insanity."
*per Prof Alexie Efros (UC Berkeley), GAN (generative adversarial network, the most sophisticated algo and used for good in autonomous driving, ugly in creating more Max Headroom, bad in deepfake video) is 3 months, so I'd put the rest of ML probably 6 months. Open source stuff happens fast.
0
You are not authorized to post a reply.
Technical Interest Groups
--Carbon Capture Utilization and...
--Deep Learning - Machine Learni...
--Detrital Zircon TIG Discussion...
--Drones and Drone Data TIG Disc...
--Geomechanics and Hydraulic Fra...
--Petrophysics TIG Discussions
--Playmaker to Bulletin Article ...
--Produced Water Solutions TIG D...
--Unconventional Research TIG Di...
--Stratigraphy and Sedimentology...
--Triggered and Induced Seismici...
--Salt Basins TIG Discussions
--Geoscience Technicians and GIS...
--Petroleum Systems TIG Discussi...
Special Interest Group
--Jammin' GEOS SIG Discussions
--AAPG Women's Network SIG Discu...
--Visiting Geoscientist SIG Disc...
--Young Professionals SIG Discus...
--History of Petroleum Geology S...
Forums
Technical Interest Groups
Deep Learning - Machine Learning TIG Discussions
See Also: Explorer Emphasis Article
Analyst: Uncertainty Is the Oil Industry's Greatest Challenge
See Also: Explorer Geophysical Corner
Convolutional Neural Network for Fault Prediction
Data Conditioning Tools for Preservation of Structure and Stratigraphy
Two Data Reduction Schemes: A Shallow Oil Field Example
See Also: Learn! Blog
Success with Difficult Unconventionals GTW
Navigation
About this TIG
Participants
TIG Discussions Board
Events
Activity & Reports
N.E.T. Pages
Technical Interest Groups (TIGs)
Special Interest Groups (SIGs)
Discussion Boards FAQ