BEGIN:VCALENDAR VERSION:2.0 METHOD:PUBLISH PRODID:-//Tendenci - The Open Source AMS for Associations//Tendenci Codeba se MIMEDIR//EN BEGIN:VEVENT DESCRIPTION:--- This iCal file does *NOT* confirm registration.\r\nEvent d etails subject to change. ---\r\nhttps://www.spegcs.org/events/4050/\r\n\r \nEvent Title: Domain meets Deep Neural Networks: Hybrid Physics-based Hac kathon for Geoscientists and Engineers\r\nStart Date / Time: Jul 19, 2018 07:00 AM US/Central\r\nLocation: N/A\r\nGoogle\r\nhttp://maps.google.com/m aps?q=5555+San+Felipe+Street +,Houston,Texas,77056\r\n\r\nForecast\nhttp: //www.weather.com/weather/monthly/77056\r\n\r\n \r\nHalliburton and AAPG a re collaborating to bring to AAPG community the very first Hybrid Physics- based deep neuralnetwork (DNN) Hackathon. Physics informed DNN uniquely co mbines physics and machine-learning. \r\nThe objective of this hackathon, designed for domain geoscientists and engineers, is to hack the provided P ython®\; code and develop DNN models for time series predictions of res ervoir behavior.\r\nAfter we sell out, please sign up on the waiting list -- we will open more seats.\r\nDownload Brochure\r\nHybrid Physics/DNN Int egration The use of deep learning models in oil and gas is on the rise. De ep learning models while showing a lot of promise have limitations when ap plied to oil and gas problemsSpecifically these limitations deal with bein g able to incorporate geoscientists&rsquo\; understanding of subsurface ph ysics into deep learning models\r\nSubsurface physics can be incorporated in one of many ways a subset being:Augmentation of training datasets for d eep learning models using data generated by physics driven simulators, inc luding physics-based models as a component in an ensemble of data driven d eep learning modelsFormally incorporating domain physics within the deep n eural network structure\r\nScope and PrerequisitesDomain meets DNN hackath on is designed for practitioners, data scientists, developers, decision-ma kers and includes hands-on Python coding experiments. OpenEarth&trade\; Co mmunity (OEC) , a cloud-based environment will be provided. For this hacka thon, participants will focus on\r\n\r\nAugmenting provisioned training da taset using a relevant physics based model \r\nModifying and applying supp lied algorithm to their own model \r\n\r\nREQUIRED BOOTCAMP WEBINAR/ July 16. All participants must participate in the pre-event familiarization boo t-camp webinar. The bootcamp will orient participants with Python-coding a nd OEC environment. You will receive log-in information and time.\r\nAgend a7.00 am &ndash\; Doors open 7.45 am &ndash\; Welcome and Introduction, Su san Nash, Director of Innovation, Science, and Technology, AAPG / Patrick Ng, AAPG Deep Learning TIG / Rekha Patel, Halliburton\r\n8.15am &ndash\; H ackathon Overview &ndash\; Steve Ward, Chief Advisor8.30am Hackathon Begin s &ndash\; Srinath Madasu, Technical Advisor9.30 am Coding for the Scope 1 Starts12.00 &ndash\; Lunch\r\n1.00 pm &ndash\; Coding for Scope 2 Starts3 .30 pm &ndash\; Presentations/Judging5:00 pm &ndash\; Awards &\; Networ king --- This iCal file does *NOT* confirm registration.Event details subj ect to change. ---\r\n\r\n--- By Tendenci - The Open Source AMS for Associ ations ---\r\n UID:uid4050@spegcs.org SUMMARY:Domain meets Deep Neural Networks: Hybrid Physics-based Hackathon for Geoscientists and Engineers DTSTART:20180719T120000Z DTEND:20180719T223000Z CLASS:PUBLIC PRIORITY:5 DTSTAMP:20240329T043853Z TRANSP:OPAQUE SEQUENCE:0 LOCATION:N/A X-ALT-DESC;FMTTYPE=text/html:
 \;
Hallibu rton \;and \;AAPG \; are collaborating to bring to AAPG community the very first Hybrid Physics -based deep neuralnetwork (DNN) Hackathon. Physics informed DNN uniquely c ombines physics and machine-learning. \;
The objective of this hackathon, designed for domain geoscientists and engineers, is to hack the provided Python®\; code and develop DNN models for time series predict ions of reservoir behavior.
After we sell out, please sign up on the waiting list -- we will open more seats.
Hybrid Physics/DNN Integration \;
The use of de
ep learning models in oil and gas is on the rise. \;
Deep learnin
g models while showing a lot of promise have limitations when applied to o
il and gas problems
Specifically these limitations deal with being ab
le to incorporate geoscientists&rsquo\; understanding of subsurface physic
s into deep learning models
Subsurface physics can be incorporated
in one of many ways a subset being:
Augmentation of training datasets
for deep learning models using data generated by physics driven simulator
s, including physics-based models as a component in an ensemble of data dr
iven deep learning models
Formally incorporating domain physics withi
n the deep neural network structure
Scope and Prerequisites
Domain meets DNN hackathon is designed for practitioners, d
ata scientists, developers, decision-makers and includes hands-on Python c
oding experiments. OpenEarth&trade\; Community (OEC) , a cloud-based envir
onment will be provided. For this hackathon, participants will focus on
REQUIRED BOOTCAMP WEBINAR/ July 16. All participants must participate in the pre-event familiarization boot-ca mp webinar. The bootcamp will orient participants with Python-coding and O EC environment. You will receive log-in information and time.
7.00 am &ndash\; Doors open \;
7.45 am &n
dash\; Welcome and Introduction, Susan Nash, Director of Innovation, Scien
ce, and Technology, AAPG / Patrick Ng, AAPG Deep Learning TIG / Rekha Pate
l, Halliburton
8.15am &ndash\; Hackathon Overview &ndash\; Steve Wa
rd, Chief Advisor
8.30am Hackathon Begins &ndash\; Srinath Madasu, Te
chnical Advisor
9.30 am Coding for the Scope 1 Starts
12.00 &nda
sh\; Lunch
1.00 pm &ndash\; Coding for Scope 2 Starts
3.30 pm
&ndash\; Presentations/Judging
5:00 pm &ndash\; Awards &\; Network
ing \;