SlideShare une entreprise Scribd logo
1  sur  34
Learning Systems for Science
Ian Foster
Argonne National Laboratory and The University of Chicago
foster@anl.gov
1
Joint work with Rachana Ananthakrishnan, Ben Blaiszik, Kyle Chard, Ryan Chard,
Mike Papka, Jim Pruyne, Steve Tuecke, Rick Wagner, Logan Ward, and others
“Whatever you are studying right now if
you are not getting up to speed on deep
learning, neural networks, etc., you lose.
We are going through the process where
software will automate software,
automation will automate automation.”
-- Mark Cuban
Deep leaning is also finding applications in science.
Example: Predicting formation enthalpies of crystalline materials
Best conventional machine learning method,
Random Forest:
a) Only elemental
compositions
(DFT-computed OQMD)
Given DHf e.g.:
Cr2Ni3
Al2O3
Predict:
TiO2 ?
Logan Ward et al.,
Phys Rev B, 2017
Best conventional machine learning method,
Random Forest:
a) Only elemental
compositions
b) Also physical
attributes
(DFT-computed OQMD)
Compute
145 physical
properties:
• Stoichiometric
• Elemental
property
statistics
• Electronic
structure
• Ionic
compound
Deep leaning is also finding applications in science.
Example: Predicting formation enthalpies of crystalline materials
Logan Ward et al.,
Phys Rev B, 2017
Best conventional machine learning method,
Random Forest:
a) Only elemental
compositions
b) Also physical
attributes
Dipendra Jha
ElemNet,
17-layer DNN:
Only elemental
compositions
(Also runs 100x
faster than RF.)
3,500550
Jha, Ward, et al., 2018.
(DFT-computed OQMD)
Deep leaning is also finding applications in science.
Example: Predicting formation enthalpies of crystalline materials
Logan Ward et al.,
Phys Rev B, 2017
Deep learning
• Drug response prediction
• Scientific image classification
• Scientific text understanding
• Materials property design
• Gravitational lens detection
• Feature detection in 3D
• Street scene analysis
• Organism design
• State space prediction
• Persistent learning
• Hyperspectral patterns
Many other interesting applications are emerging
Simulation
• Materials science
• Cosmology
• Molecular dynamics
• Nuclear reactor modeling
• Combustion
• Quantum computer
simulation
• Climate modeling
• Power grid
• Discrete event simulation
• Fusion reactor simulation
• Brain simulation
• Transportation networks
Big data
• APS data analysis
• HEP data analysis
• LSST data analysis
• SKA data analysis
• Metagenome analysis
• Battery design search
• Graph analysis
• Virtual compound library
• Neuroscience data analysis
• Genome pipelines
Rick Stevens: Argonne applications for exascale
Jack was right:
It’s linear algebra all
the way down
https://xkcd.com/1838/
We face many research challenges
9
Applications
Learning systems
Foundations
Hardware
Mathematics, algorithms; general AI, reinforcement
learning, uncertainty quantification, explanability,
etc.
Advanced hardware to support AI. Evaluation of new
architectures and systems. Neuromorphic and quantum
as long-term AI accelerators?
AI software. Software infrastructure for managing data,
models, workflows etc., and for delivering AI capabilities
to 10,000s of scientists and engineers.
AI applications across science and engineering. New
approaches to simulation and experimental science.
DeepAI
We need a lot more computing
Exaflop/s-days used to
train:
AlexNet: 0.000007
(in 2012)
AlphaGo Zero: 2
(in 2017)
x 300,000 in 5.5 years
Opportunities for science automation:
Research today
11
Configure apparatus/write code
Run experiments
Solve
societal
problems
Create knowledge
What scientists
want to do
Most
scientist
time
Analyze and plan
12
Run experiments
Create knowledge
Most
scientist
time
AI
assistants
Analyze and plan
Opportunities for science automation:
Research tomorrow
Solve
societal
problems
Configure apparatus/write code
Example: Accelerated discovery of metallic glasses
Metallic glasses offer unique
properties, but discovering
new, useful alloys is slow
• ML model predicts glass
formation
• Validate with automated
experimentation
• Active learning to optimize
experiments
13
Ren et al. Sci Adv. (2017) eaaq1566
Random forest to predict metallic glass formation
Batch active learning to choose experiments
Discovery of new ternary glass systems
14
Ren et al. Sci Adv. (2017) eaaq1566
Example: Accelerated discovery of metallic glasses
15
Imagine when
only the fun parts
of science remain
https://twitter.com/worrydream/status/992546529217933312
16
Developing a DL model remains an artisanal process
Model
selection
Model
training
Inference
Training
data
Q
A
Training
data
Human
expertise
model
architecture
trained
model
Many challenges. For example …
• Finding relevant models and methods (1000s of papers per year)
• Finding relevant data for training and validation
• Implementing, training, testing, and validating models
• Configuring and adapting models
• Scaling, accelerating, and optimizing models
• Leveraging new architectures
• Integrating models into scientific work processes
• Documenting, sharing, and explaining results
• Integrating and applying advanced methods: UQ, active learning,
reinforcement learning, …
• Engaging and educating the non-expert 99.99%
18
Learning
systems
AI software. Software infrastructure for managing
data, models, workflows etc., and for delivering AI
capabilities to 10,000s of scientists and engineers.
“Without deep understanding of the basic tools needed to build and train new
algorithms … researchers creating AIs resort to hearsay, like medieval alchemists.
People gravitate around cargo-cult practices, relying on folklore and magic spells.”
– Science, May 3 2018
New “learning systems for science”
Organizing relevant data: Materials Data Facility
EP
EP
EP
• Query
• Browse
• Aggregate
• Mint DOIs
• Associate
metadata
• Persist
datasets
Databases
Datasets
APIs
LIMS
etc.
Distributed data
storage
Data
Publication
Data
Discovery
materialsdatafacility.org
Ben Blaiszik, Logan Ward, Jonathan Gaff, and others
DLHub: A data and learning hub for science
• Collect, publish, categorize models/code/ weights/data from many sources
• Serve models via API to foster sharing, consumption, and access to data,
training sets, and models
• Automate training of models
(using HPC as needed) as
new data are available
• Enable new science through
reuse and synthesis of existing
models
TrainCollect Serve
Ben Blaiszik, Ryan Chard, Logan Ward, and others
“beam misaligned”
“…”
Say you want to use a deep neural network for online identification
of problems when running diffraction experiments
DLHub: Collect, serve, train community models
https://doi.org/10.1109/NYSDS.2017.8085045
DLHub: Collect, serve, train community models
▪ Where are the model and trained weights?
▪ How do I run the model on my data?
▪ Should I run the model on my data?
▪ How can I retrain the model on new data?
https://doi.org/10.1109/NYSDS.2017.8085045
DLHub: Collect, serve, train community models
DLHub
[“beam off image”, …]
model/xray/batch_predict
▪ Where are the model and trained weights?
▪ How do I run the model on my data?
▪ Should I run the model on my data?
▪ How can I retrain the model on new data?
https://doi.org/10.1109/NYSDS.2017.8085045
DLHub: Collect, serve, train community models
DLHub
[“beam off image”, …]
model/xray/batch_predict
▪ Where are the model and trained weights?
▪ How do I run the model on my data?
▪ Should I run the model on my data?
▪ How can I retrain the model on new data?
https://doi.org/10.1109/NYSDS.2017.8085045
DLHub: Collect, serve, train community models
DLHub
Collect
Data
1) Register a model
Train
Model
Register
Model Model /
transform
containers
Receive DOI
Send to DLHub
DLHub: Collect, serve, train community models
DLHub
Collect
Data
Receive
predicted
Properties
Send
compositions
Call
DLHub
Find
Model
2) Run a model
Model /
transform
containers
DLHub: Collect, serve, train community models
Collect
Data
Receive DOI
1) Register a model
Train
Model
Register
Model
Send to DLHub
DLHub: Initial Use Cases
• X-Ray diffraction (XRD) image tagging model
• Prediction of bulk metallic glass forming regions
in ternary diagrams
• Predicting compound stability and
bandgap by elemental composition
Coming Soon
• Deep learning to predict crystalline materials
• ML/DL applied to high-throughput catalyst
synthesis, simulation, and characterization
• DL for chemical compound stability prediction
• High-throughput High-Energy Diffraction
Microscopy (HEDM) analysis
with SLAC, NIST, NU, USC, Citrine
With
CHiMaD/NU
Wang,
Yager
et al.
globus.org
31
32
Invoke model on data
Ben Blaiszik Steve TueckeKyle Chard Jim Pruyne Logan WardRachana
Ananthakrishnan
Ryan Chard Mike Papka Rick Wagner
I reported on the work of many talented people
Thanks also to:
• Jon Almer, Francesco de Carlo, Hemant Sharma, Brian Toby, Stefan Vogt, Stephen Streiffer,
Nicholas Schwarz, Doga Gursoy, and others, Advanced Photon Source
• Tekin Bicer, Jonathan Gaff, Raj Kettimuthu, Justin Wozniak, and others, Argonne Computing
We thank our sponsors
DLHub Globus
IMaD
Petrel
Argonne Leadership
Computing Facility
34
Applications
Learning systems
Foundations
Hardware
Mathematics, algorithms; general AI, reinforcement
learning, uncertainty quantification, explanability,
etc.
Advanced hardware to support AI. Evaluation of new
architectures and systems. Neuromorphic and quantum
as long-term AI accelerators?
AI software. Software infrastructure for managing data,
models, workflows etc., and for delivering AI capabilities
to 10,000s of scientists and engineers.
AI applications across science and engineering. New
approaches to simulation and experimental science.
“All the impressive achievements of deep learning
amount to just curve fitting.” – Judea Pearl

Contenu connexe

Tendances

Materials Data Facility: Streamlined and automated data sharing, discovery, ...
Materials Data Facility: Streamlined and automated data sharing,  discovery, ...Materials Data Facility: Streamlined and automated data sharing,  discovery, ...
Materials Data Facility: Streamlined and automated data sharing, discovery, ...Ian Foster
 
The Discovery Cloud: Accelerating Science via Outsourcing and Automation
The Discovery Cloud: Accelerating Science via Outsourcing and AutomationThe Discovery Cloud: Accelerating Science via Outsourcing and Automation
The Discovery Cloud: Accelerating Science via Outsourcing and AutomationIan Foster
 
AI at Scale for Materials and Chemistry
AI at Scale for Materials and ChemistryAI at Scale for Materials and Chemistry
AI at Scale for Materials and ChemistryIan Foster
 
Open Science Data Cloud - CCA 11
Open Science Data Cloud - CCA 11Open Science Data Cloud - CCA 11
Open Science Data Cloud - CCA 11Robert Grossman
 
Open Science Data Cloud (IEEE Cloud 2011)
Open Science Data Cloud (IEEE Cloud 2011)Open Science Data Cloud (IEEE Cloud 2011)
Open Science Data Cloud (IEEE Cloud 2011)Robert Grossman
 
Share and analyze geonomic data at scale by Andy Petrella and Xavier Tordoir
Share and analyze geonomic data at scale by Andy Petrella and Xavier TordoirShare and analyze geonomic data at scale by Andy Petrella and Xavier Tordoir
Share and analyze geonomic data at scale by Andy Petrella and Xavier TordoirSpark Summit
 
Accelerated Materials Discovery Using Theory, Optimization, and Natural Langu...
Accelerated Materials Discovery Using Theory, Optimization, and Natural Langu...Accelerated Materials Discovery Using Theory, Optimization, and Natural Langu...
Accelerated Materials Discovery Using Theory, Optimization, and Natural Langu...Anubhav Jain
 
NERSC, AI and the Superfacility, Debbie Bard
NERSC, AI and the Superfacility, Debbie BardNERSC, AI and the Superfacility, Debbie Bard
NERSC, AI and the Superfacility, Debbie BardPacificResearchPlatform
 
Accelerating Data-driven Discovery in Energy Science
Accelerating Data-driven Discovery in Energy ScienceAccelerating Data-driven Discovery in Energy Science
Accelerating Data-driven Discovery in Energy ScienceIan Foster
 
Cloud com foster december 2010
Cloud com foster december 2010Cloud com foster december 2010
Cloud com foster december 2010Ian Foster
 
Accelerating data-intensive science by outsourcing the mundane
Accelerating data-intensive science by outsourcing the mundaneAccelerating data-intensive science by outsourcing the mundane
Accelerating data-intensive science by outsourcing the mundaneIan Foster
 
Accelerating the Experimental Feedback Loop: Data Streams and the Advanced Ph...
Accelerating the Experimental Feedback Loop: Data Streams and the Advanced Ph...Accelerating the Experimental Feedback Loop: Data Streams and the Advanced Ph...
Accelerating the Experimental Feedback Loop: Data Streams and the Advanced Ph...Ian Foster
 
Bionimbus Cambridge Workshop (3-28-11, v7)
Bionimbus Cambridge Workshop (3-28-11, v7)Bionimbus Cambridge Workshop (3-28-11, v7)
Bionimbus Cambridge Workshop (3-28-11, v7)Robert Grossman
 
The Open Science Data Cloud: Empowering the Long Tail of Science
The Open Science Data Cloud: Empowering the Long Tail of ScienceThe Open Science Data Cloud: Empowering the Long Tail of Science
The Open Science Data Cloud: Empowering the Long Tail of ScienceRobert Grossman
 
The DuraMat Data Hub and Analytics Capability: A Resource for Solar PV Data
The DuraMat Data Hub and Analytics Capability: A Resource for Solar PV DataThe DuraMat Data Hub and Analytics Capability: A Resource for Solar PV Data
The DuraMat Data Hub and Analytics Capability: A Resource for Solar PV DataAnubhav Jain
 
DuraMat Data Management and Analytics
DuraMat Data Management and AnalyticsDuraMat Data Management and Analytics
DuraMat Data Management and AnalyticsAnubhav Jain
 
What Are Science Clouds?
What Are Science Clouds?What Are Science Clouds?
What Are Science Clouds?Robert Grossman
 
Stanford/SLAC Cryo-EM Computing and Storage, Yee-Ting Li
Stanford/SLAC Cryo-EM Computing and Storage, Yee-Ting LiStanford/SLAC Cryo-EM Computing and Storage, Yee-Ting Li
Stanford/SLAC Cryo-EM Computing and Storage, Yee-Ting LiPacificResearchPlatform
 
An Overview of Bionimbus (March 2010)
An Overview of Bionimbus (March 2010)An Overview of Bionimbus (March 2010)
An Overview of Bionimbus (March 2010)Robert Grossman
 

Tendances (20)

Materials Data Facility: Streamlined and automated data sharing, discovery, ...
Materials Data Facility: Streamlined and automated data sharing,  discovery, ...Materials Data Facility: Streamlined and automated data sharing,  discovery, ...
Materials Data Facility: Streamlined and automated data sharing, discovery, ...
 
The Discovery Cloud: Accelerating Science via Outsourcing and Automation
The Discovery Cloud: Accelerating Science via Outsourcing and AutomationThe Discovery Cloud: Accelerating Science via Outsourcing and Automation
The Discovery Cloud: Accelerating Science via Outsourcing and Automation
 
AI at Scale for Materials and Chemistry
AI at Scale for Materials and ChemistryAI at Scale for Materials and Chemistry
AI at Scale for Materials and Chemistry
 
Open Science Data Cloud - CCA 11
Open Science Data Cloud - CCA 11Open Science Data Cloud - CCA 11
Open Science Data Cloud - CCA 11
 
Open Science Data Cloud (IEEE Cloud 2011)
Open Science Data Cloud (IEEE Cloud 2011)Open Science Data Cloud (IEEE Cloud 2011)
Open Science Data Cloud (IEEE Cloud 2011)
 
Share and analyze geonomic data at scale by Andy Petrella and Xavier Tordoir
Share and analyze geonomic data at scale by Andy Petrella and Xavier TordoirShare and analyze geonomic data at scale by Andy Petrella and Xavier Tordoir
Share and analyze geonomic data at scale by Andy Petrella and Xavier Tordoir
 
Accelerated Materials Discovery Using Theory, Optimization, and Natural Langu...
Accelerated Materials Discovery Using Theory, Optimization, and Natural Langu...Accelerated Materials Discovery Using Theory, Optimization, and Natural Langu...
Accelerated Materials Discovery Using Theory, Optimization, and Natural Langu...
 
ML in materials discovery
ML in materials discovery ML in materials discovery
ML in materials discovery
 
NERSC, AI and the Superfacility, Debbie Bard
NERSC, AI and the Superfacility, Debbie BardNERSC, AI and the Superfacility, Debbie Bard
NERSC, AI and the Superfacility, Debbie Bard
 
Accelerating Data-driven Discovery in Energy Science
Accelerating Data-driven Discovery in Energy ScienceAccelerating Data-driven Discovery in Energy Science
Accelerating Data-driven Discovery in Energy Science
 
Cloud com foster december 2010
Cloud com foster december 2010Cloud com foster december 2010
Cloud com foster december 2010
 
Accelerating data-intensive science by outsourcing the mundane
Accelerating data-intensive science by outsourcing the mundaneAccelerating data-intensive science by outsourcing the mundane
Accelerating data-intensive science by outsourcing the mundane
 
Accelerating the Experimental Feedback Loop: Data Streams and the Advanced Ph...
Accelerating the Experimental Feedback Loop: Data Streams and the Advanced Ph...Accelerating the Experimental Feedback Loop: Data Streams and the Advanced Ph...
Accelerating the Experimental Feedback Loop: Data Streams and the Advanced Ph...
 
Bionimbus Cambridge Workshop (3-28-11, v7)
Bionimbus Cambridge Workshop (3-28-11, v7)Bionimbus Cambridge Workshop (3-28-11, v7)
Bionimbus Cambridge Workshop (3-28-11, v7)
 
The Open Science Data Cloud: Empowering the Long Tail of Science
The Open Science Data Cloud: Empowering the Long Tail of ScienceThe Open Science Data Cloud: Empowering the Long Tail of Science
The Open Science Data Cloud: Empowering the Long Tail of Science
 
The DuraMat Data Hub and Analytics Capability: A Resource for Solar PV Data
The DuraMat Data Hub and Analytics Capability: A Resource for Solar PV DataThe DuraMat Data Hub and Analytics Capability: A Resource for Solar PV Data
The DuraMat Data Hub and Analytics Capability: A Resource for Solar PV Data
 
DuraMat Data Management and Analytics
DuraMat Data Management and AnalyticsDuraMat Data Management and Analytics
DuraMat Data Management and Analytics
 
What Are Science Clouds?
What Are Science Clouds?What Are Science Clouds?
What Are Science Clouds?
 
Stanford/SLAC Cryo-EM Computing and Storage, Yee-Ting Li
Stanford/SLAC Cryo-EM Computing and Storage, Yee-Ting LiStanford/SLAC Cryo-EM Computing and Storage, Yee-Ting Li
Stanford/SLAC Cryo-EM Computing and Storage, Yee-Ting Li
 
An Overview of Bionimbus (March 2010)
An Overview of Bionimbus (March 2010)An Overview of Bionimbus (March 2010)
An Overview of Bionimbus (March 2010)
 

Similaire à Deep Learning Applications in Science

Foundations for the Future of Science
Foundations for the Future of ScienceFoundations for the Future of Science
Foundations for the Future of ScienceGlobus
 
Discovering new functional materials for clean energy and beyond using high-t...
Discovering new functional materials for clean energy and beyond using high-t...Discovering new functional materials for clean energy and beyond using high-t...
Discovering new functional materials for clean energy and beyond using high-t...Anubhav Jain
 
A Data Ecosystem to Support Machine Learning in Materials Science
A Data Ecosystem to Support Machine Learning in Materials ScienceA Data Ecosystem to Support Machine Learning in Materials Science
A Data Ecosystem to Support Machine Learning in Materials ScienceGlobus
 
Materials discovery through theory, computation, and machine learning
Materials discovery through theory, computation, and machine learningMaterials discovery through theory, computation, and machine learning
Materials discovery through theory, computation, and machine learningAnubhav Jain
 
eScience: A Transformed Scientific Method
eScience: A Transformed Scientific MethodeScience: A Transformed Scientific Method
eScience: A Transformed Scientific MethodDuncan Hull
 
Discovery Engines for Big Data: Accelerating Discovery in Basic Energy Sciences
Discovery Engines for Big Data: Accelerating Discovery in Basic Energy SciencesDiscovery Engines for Big Data: Accelerating Discovery in Basic Energy Sciences
Discovery Engines for Big Data: Accelerating Discovery in Basic Energy SciencesIan Foster
 
Hattrick-Simpers MRS Webinar on AI in Materials
Hattrick-Simpers MRS Webinar on AI in MaterialsHattrick-Simpers MRS Webinar on AI in Materials
Hattrick-Simpers MRS Webinar on AI in MaterialsJason Hattrick-Simpers
 
Time to Science/Time to Results: Transforming Research in the Cloud
Time to Science/Time to Results: Transforming Research in the CloudTime to Science/Time to Results: Transforming Research in the Cloud
Time to Science/Time to Results: Transforming Research in the CloudAmazon Web Services
 
H2O with Erin LeDell at Portland R User Group
H2O with Erin LeDell at Portland R User GroupH2O with Erin LeDell at Portland R User Group
H2O with Erin LeDell at Portland R User GroupSri Ambati
 
Computation and Knowledge
Computation and KnowledgeComputation and Knowledge
Computation and KnowledgeIan Foster
 
Deep Learning on Apache Spark at CERN’s Large Hadron Collider with Intel Tech...
Deep Learning on Apache Spark at CERN’s Large Hadron Collider with Intel Tech...Deep Learning on Apache Spark at CERN’s Large Hadron Collider with Intel Tech...
Deep Learning on Apache Spark at CERN’s Large Hadron Collider with Intel Tech...Databricks
 
Physics inspired artificial intelligence/machine learning
Physics inspired artificial intelligence/machine learningPhysics inspired artificial intelligence/machine learning
Physics inspired artificial intelligence/machine learningKAMAL CHOUDHARY
 
2D/3D Materials screening and genetic algorithm with ML model
2D/3D Materials screening and genetic algorithm with ML model2D/3D Materials screening and genetic algorithm with ML model
2D/3D Materials screening and genetic algorithm with ML modelaimsnist
 
Materials Data in the 21st Century: From Mishmash to Moneyball
Materials Data in the 21st Century: From Mishmash to MoneyballMaterials Data in the 21st Century: From Mishmash to Moneyball
Materials Data in the 21st Century: From Mishmash to Moneyballbmeredig
 
The Materials Data Facility: A Distributed Model for the Materials Data Commu...
The Materials Data Facility: A Distributed Model for the Materials Data Commu...The Materials Data Facility: A Distributed Model for the Materials Data Commu...
The Materials Data Facility: A Distributed Model for the Materials Data Commu...Ben Blaiszik
 
Data Mining to Discovery for Inorganic Solids: Software Tools and Applications
Data Mining to Discovery for Inorganic Solids: Software Tools and ApplicationsData Mining to Discovery for Inorganic Solids: Software Tools and Applications
Data Mining to Discovery for Inorganic Solids: Software Tools and Applicationsaimsnist
 
Data-knowledge transition zones within the biomedical research ecosystem
Data-knowledge transition zones within the biomedical research ecosystemData-knowledge transition zones within the biomedical research ecosystem
Data-knowledge transition zones within the biomedical research ecosystemMaryann Martone
 

Similaire à Deep Learning Applications in Science (20)

Foundations for the Future of Science
Foundations for the Future of ScienceFoundations for the Future of Science
Foundations for the Future of Science
 
Discovering new functional materials for clean energy and beyond using high-t...
Discovering new functional materials for clean energy and beyond using high-t...Discovering new functional materials for clean energy and beyond using high-t...
Discovering new functional materials for clean energy and beyond using high-t...
 
A Data Ecosystem to Support Machine Learning in Materials Science
A Data Ecosystem to Support Machine Learning in Materials ScienceA Data Ecosystem to Support Machine Learning in Materials Science
A Data Ecosystem to Support Machine Learning in Materials Science
 
Materials discovery through theory, computation, and machine learning
Materials discovery through theory, computation, and machine learningMaterials discovery through theory, computation, and machine learning
Materials discovery through theory, computation, and machine learning
 
AI for Science
AI for ScienceAI for Science
AI for Science
 
eScience: A Transformed Scientific Method
eScience: A Transformed Scientific MethodeScience: A Transformed Scientific Method
eScience: A Transformed Scientific Method
 
Discovery Engines for Big Data: Accelerating Discovery in Basic Energy Sciences
Discovery Engines for Big Data: Accelerating Discovery in Basic Energy SciencesDiscovery Engines for Big Data: Accelerating Discovery in Basic Energy Sciences
Discovery Engines for Big Data: Accelerating Discovery in Basic Energy Sciences
 
Hattrick-Simpers MRS Webinar on AI in Materials
Hattrick-Simpers MRS Webinar on AI in MaterialsHattrick-Simpers MRS Webinar on AI in Materials
Hattrick-Simpers MRS Webinar on AI in Materials
 
Time to Science/Time to Results: Transforming Research in the Cloud
Time to Science/Time to Results: Transforming Research in the CloudTime to Science/Time to Results: Transforming Research in the Cloud
Time to Science/Time to Results: Transforming Research in the Cloud
 
H2O with Erin LeDell at Portland R User Group
H2O with Erin LeDell at Portland R User GroupH2O with Erin LeDell at Portland R User Group
H2O with Erin LeDell at Portland R User Group
 
Computation and Knowledge
Computation and KnowledgeComputation and Knowledge
Computation and Knowledge
 
Deep Learning on Apache Spark at CERN’s Large Hadron Collider with Intel Tech...
Deep Learning on Apache Spark at CERN’s Large Hadron Collider with Intel Tech...Deep Learning on Apache Spark at CERN’s Large Hadron Collider with Intel Tech...
Deep Learning on Apache Spark at CERN’s Large Hadron Collider with Intel Tech...
 
Big Data
Big Data Big Data
Big Data
 
Physics inspired artificial intelligence/machine learning
Physics inspired artificial intelligence/machine learningPhysics inspired artificial intelligence/machine learning
Physics inspired artificial intelligence/machine learning
 
Summary of 3DPAS
Summary of 3DPASSummary of 3DPAS
Summary of 3DPAS
 
2D/3D Materials screening and genetic algorithm with ML model
2D/3D Materials screening and genetic algorithm with ML model2D/3D Materials screening and genetic algorithm with ML model
2D/3D Materials screening and genetic algorithm with ML model
 
Materials Data in the 21st Century: From Mishmash to Moneyball
Materials Data in the 21st Century: From Mishmash to MoneyballMaterials Data in the 21st Century: From Mishmash to Moneyball
Materials Data in the 21st Century: From Mishmash to Moneyball
 
The Materials Data Facility: A Distributed Model for the Materials Data Commu...
The Materials Data Facility: A Distributed Model for the Materials Data Commu...The Materials Data Facility: A Distributed Model for the Materials Data Commu...
The Materials Data Facility: A Distributed Model for the Materials Data Commu...
 
Data Mining to Discovery for Inorganic Solids: Software Tools and Applications
Data Mining to Discovery for Inorganic Solids: Software Tools and ApplicationsData Mining to Discovery for Inorganic Solids: Software Tools and Applications
Data Mining to Discovery for Inorganic Solids: Software Tools and Applications
 
Data-knowledge transition zones within the biomedical research ecosystem
Data-knowledge transition zones within the biomedical research ecosystemData-knowledge transition zones within the biomedical research ecosystem
Data-knowledge transition zones within the biomedical research ecosystem
 

Plus de Ian Foster

Global Services for Global Science March 2023.pptx
Global Services for Global Science March 2023.pptxGlobal Services for Global Science March 2023.pptx
Global Services for Global Science March 2023.pptxIan Foster
 
The Earth System Grid Federation: Origins, Current State, Evolution
The Earth System Grid Federation: Origins, Current State, EvolutionThe Earth System Grid Federation: Origins, Current State, Evolution
The Earth System Grid Federation: Origins, Current State, EvolutionIan Foster
 
Better Information Faster: Programming the Continuum
Better Information Faster: Programming the ContinuumBetter Information Faster: Programming the Continuum
Better Information Faster: Programming the ContinuumIan Foster
 
ESnet6 and Smart Instruments
ESnet6 and Smart InstrumentsESnet6 and Smart Instruments
ESnet6 and Smart InstrumentsIan Foster
 
Linking Scientific Instruments and Computation
Linking Scientific Instruments and ComputationLinking Scientific Instruments and Computation
Linking Scientific Instruments and ComputationIan Foster
 
A Global Research Data Platform: How Globus Services Enable Scientific Discovery
A Global Research Data Platform: How Globus Services Enable Scientific DiscoveryA Global Research Data Platform: How Globus Services Enable Scientific Discovery
A Global Research Data Platform: How Globus Services Enable Scientific DiscoveryIan Foster
 
Foster CRA March 2022.pptx
Foster CRA March 2022.pptxFoster CRA March 2022.pptx
Foster CRA March 2022.pptxIan Foster
 
Big Data, Big Computing, AI, and Environmental Science
Big Data, Big Computing, AI, and Environmental ScienceBig Data, Big Computing, AI, and Environmental Science
Big Data, Big Computing, AI, and Environmental ScienceIan Foster
 
Research Automation for Data-Driven Discovery
Research Automation for Data-Driven DiscoveryResearch Automation for Data-Driven Discovery
Research Automation for Data-Driven DiscoveryIan Foster
 
Team Argon Summary
Team Argon SummaryTeam Argon Summary
Team Argon SummaryIan Foster
 
Thoughts on interoperability
Thoughts on interoperabilityThoughts on interoperability
Thoughts on interoperabilityIan Foster
 
NIH Data Commons Architecture Ideas
NIH Data Commons Architecture IdeasNIH Data Commons Architecture Ideas
NIH Data Commons Architecture IdeasIan Foster
 
Going Smart and Deep on Materials at ALCF
Going Smart and Deep on Materials at ALCFGoing Smart and Deep on Materials at ALCF
Going Smart and Deep on Materials at ALCFIan Foster
 
Computing Just What You Need: Online Data Analysis and Reduction at Extreme ...
Computing Just What You Need: Online Data Analysis and Reduction  at Extreme ...Computing Just What You Need: Online Data Analysis and Reduction  at Extreme ...
Computing Just What You Need: Online Data Analysis and Reduction at Extreme ...Ian Foster
 
Software Infrastructure for a National Research Platform
Software Infrastructure for a National Research PlatformSoftware Infrastructure for a National Research Platform
Software Infrastructure for a National Research PlatformIan Foster
 
Globus Auth: A Research Identity and Access Management Platform
Globus Auth: A Research Identity and Access Management PlatformGlobus Auth: A Research Identity and Access Management Platform
Globus Auth: A Research Identity and Access Management PlatformIan Foster
 
Streamlined data sharing and analysis to accelerate cancer research
Streamlined data sharing and analysis to accelerate cancer researchStreamlined data sharing and analysis to accelerate cancer research
Streamlined data sharing and analysis to accelerate cancer researchIan Foster
 
Science Services and Science Platforms: Using the Cloud to Accelerate and Dem...
Science Services and Science Platforms: Using the Cloud to Accelerate and Dem...Science Services and Science Platforms: Using the Cloud to Accelerate and Dem...
Science Services and Science Platforms: Using the Cloud to Accelerate and Dem...Ian Foster
 
building global software/earthcube->sciencecloud
building global software/earthcube->sciencecloudbuilding global software/earthcube->sciencecloud
building global software/earthcube->sciencecloudIan Foster
 

Plus de Ian Foster (19)

Global Services for Global Science March 2023.pptx
Global Services for Global Science March 2023.pptxGlobal Services for Global Science March 2023.pptx
Global Services for Global Science March 2023.pptx
 
The Earth System Grid Federation: Origins, Current State, Evolution
The Earth System Grid Federation: Origins, Current State, EvolutionThe Earth System Grid Federation: Origins, Current State, Evolution
The Earth System Grid Federation: Origins, Current State, Evolution
 
Better Information Faster: Programming the Continuum
Better Information Faster: Programming the ContinuumBetter Information Faster: Programming the Continuum
Better Information Faster: Programming the Continuum
 
ESnet6 and Smart Instruments
ESnet6 and Smart InstrumentsESnet6 and Smart Instruments
ESnet6 and Smart Instruments
 
Linking Scientific Instruments and Computation
Linking Scientific Instruments and ComputationLinking Scientific Instruments and Computation
Linking Scientific Instruments and Computation
 
A Global Research Data Platform: How Globus Services Enable Scientific Discovery
A Global Research Data Platform: How Globus Services Enable Scientific DiscoveryA Global Research Data Platform: How Globus Services Enable Scientific Discovery
A Global Research Data Platform: How Globus Services Enable Scientific Discovery
 
Foster CRA March 2022.pptx
Foster CRA March 2022.pptxFoster CRA March 2022.pptx
Foster CRA March 2022.pptx
 
Big Data, Big Computing, AI, and Environmental Science
Big Data, Big Computing, AI, and Environmental ScienceBig Data, Big Computing, AI, and Environmental Science
Big Data, Big Computing, AI, and Environmental Science
 
Research Automation for Data-Driven Discovery
Research Automation for Data-Driven DiscoveryResearch Automation for Data-Driven Discovery
Research Automation for Data-Driven Discovery
 
Team Argon Summary
Team Argon SummaryTeam Argon Summary
Team Argon Summary
 
Thoughts on interoperability
Thoughts on interoperabilityThoughts on interoperability
Thoughts on interoperability
 
NIH Data Commons Architecture Ideas
NIH Data Commons Architecture IdeasNIH Data Commons Architecture Ideas
NIH Data Commons Architecture Ideas
 
Going Smart and Deep on Materials at ALCF
Going Smart and Deep on Materials at ALCFGoing Smart and Deep on Materials at ALCF
Going Smart and Deep on Materials at ALCF
 
Computing Just What You Need: Online Data Analysis and Reduction at Extreme ...
Computing Just What You Need: Online Data Analysis and Reduction  at Extreme ...Computing Just What You Need: Online Data Analysis and Reduction  at Extreme ...
Computing Just What You Need: Online Data Analysis and Reduction at Extreme ...
 
Software Infrastructure for a National Research Platform
Software Infrastructure for a National Research PlatformSoftware Infrastructure for a National Research Platform
Software Infrastructure for a National Research Platform
 
Globus Auth: A Research Identity and Access Management Platform
Globus Auth: A Research Identity and Access Management PlatformGlobus Auth: A Research Identity and Access Management Platform
Globus Auth: A Research Identity and Access Management Platform
 
Streamlined data sharing and analysis to accelerate cancer research
Streamlined data sharing and analysis to accelerate cancer researchStreamlined data sharing and analysis to accelerate cancer research
Streamlined data sharing and analysis to accelerate cancer research
 
Science Services and Science Platforms: Using the Cloud to Accelerate and Dem...
Science Services and Science Platforms: Using the Cloud to Accelerate and Dem...Science Services and Science Platforms: Using the Cloud to Accelerate and Dem...
Science Services and Science Platforms: Using the Cloud to Accelerate and Dem...
 
building global software/earthcube->sciencecloud
building global software/earthcube->sciencecloudbuilding global software/earthcube->sciencecloud
building global software/earthcube->sciencecloud
 

Dernier

Animal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxAnimal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxUmerFayaz5
 
Traditional Agroforestry System in India- Shifting Cultivation, Taungya, Home...
Traditional Agroforestry System in India- Shifting Cultivation, Taungya, Home...Traditional Agroforestry System in India- Shifting Cultivation, Taungya, Home...
Traditional Agroforestry System in India- Shifting Cultivation, Taungya, Home...jana861314
 
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 60009654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000Sapana Sha
 
Nanoparticles synthesis and characterization​ ​
Nanoparticles synthesis and characterization​  ​Nanoparticles synthesis and characterization​  ​
Nanoparticles synthesis and characterization​ ​kaibalyasahoo82800
 
Botany krishna series 2nd semester Only Mcq type questions
Botany krishna series 2nd semester Only Mcq type questionsBotany krishna series 2nd semester Only Mcq type questions
Botany krishna series 2nd semester Only Mcq type questionsSumit Kumar yadav
 
Orientation, design and principles of polyhouse
Orientation, design and principles of polyhouseOrientation, design and principles of polyhouse
Orientation, design and principles of polyhousejana861314
 
Presentation Vikram Lander by Vedansh Gupta.pptx
Presentation Vikram Lander by Vedansh Gupta.pptxPresentation Vikram Lander by Vedansh Gupta.pptx
Presentation Vikram Lander by Vedansh Gupta.pptxgindu3009
 
Chromatin Structure | EUCHROMATIN | HETEROCHROMATIN
Chromatin Structure | EUCHROMATIN | HETEROCHROMATINChromatin Structure | EUCHROMATIN | HETEROCHROMATIN
Chromatin Structure | EUCHROMATIN | HETEROCHROMATINsankalpkumarsahoo174
 
Disentangling the origin of chemical differences using GHOST
Disentangling the origin of chemical differences using GHOSTDisentangling the origin of chemical differences using GHOST
Disentangling the origin of chemical differences using GHOSTSérgio Sacani
 
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls AgencyHire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls AgencySheetal Arora
 
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCESTERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCEPRINCE C P
 
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bNightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bSérgio Sacani
 
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...Sérgio Sacani
 
Biopesticide (2).pptx .This slides helps to know the different types of biop...
Biopesticide (2).pptx  .This slides helps to know the different types of biop...Biopesticide (2).pptx  .This slides helps to know the different types of biop...
Biopesticide (2).pptx .This slides helps to know the different types of biop...RohitNehra6
 
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPirithiRaju
 
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptxSOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptxkessiyaTpeter
 
DIFFERENCE IN BACK CROSS AND TEST CROSS
DIFFERENCE IN  BACK CROSS AND TEST CROSSDIFFERENCE IN  BACK CROSS AND TEST CROSS
DIFFERENCE IN BACK CROSS AND TEST CROSSLeenakshiTyagi
 
Green chemistry and Sustainable development.pptx
Green chemistry  and Sustainable development.pptxGreen chemistry  and Sustainable development.pptx
Green chemistry and Sustainable development.pptxRajatChauhan518211
 
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...Sérgio Sacani
 
GFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxGFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxAleenaTreesaSaji
 

Dernier (20)

Animal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxAnimal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptx
 
Traditional Agroforestry System in India- Shifting Cultivation, Taungya, Home...
Traditional Agroforestry System in India- Shifting Cultivation, Taungya, Home...Traditional Agroforestry System in India- Shifting Cultivation, Taungya, Home...
Traditional Agroforestry System in India- Shifting Cultivation, Taungya, Home...
 
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 60009654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
9654467111 Call Girls In Raj Nagar Delhi Short 1500 Night 6000
 
Nanoparticles synthesis and characterization​ ​
Nanoparticles synthesis and characterization​  ​Nanoparticles synthesis and characterization​  ​
Nanoparticles synthesis and characterization​ ​
 
Botany krishna series 2nd semester Only Mcq type questions
Botany krishna series 2nd semester Only Mcq type questionsBotany krishna series 2nd semester Only Mcq type questions
Botany krishna series 2nd semester Only Mcq type questions
 
Orientation, design and principles of polyhouse
Orientation, design and principles of polyhouseOrientation, design and principles of polyhouse
Orientation, design and principles of polyhouse
 
Presentation Vikram Lander by Vedansh Gupta.pptx
Presentation Vikram Lander by Vedansh Gupta.pptxPresentation Vikram Lander by Vedansh Gupta.pptx
Presentation Vikram Lander by Vedansh Gupta.pptx
 
Chromatin Structure | EUCHROMATIN | HETEROCHROMATIN
Chromatin Structure | EUCHROMATIN | HETEROCHROMATINChromatin Structure | EUCHROMATIN | HETEROCHROMATIN
Chromatin Structure | EUCHROMATIN | HETEROCHROMATIN
 
Disentangling the origin of chemical differences using GHOST
Disentangling the origin of chemical differences using GHOSTDisentangling the origin of chemical differences using GHOST
Disentangling the origin of chemical differences using GHOST
 
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls AgencyHire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
 
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCESTERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
 
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bNightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
 
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
 
Biopesticide (2).pptx .This slides helps to know the different types of biop...
Biopesticide (2).pptx  .This slides helps to know the different types of biop...Biopesticide (2).pptx  .This slides helps to know the different types of biop...
Biopesticide (2).pptx .This slides helps to know the different types of biop...
 
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
 
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptxSOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
 
DIFFERENCE IN BACK CROSS AND TEST CROSS
DIFFERENCE IN  BACK CROSS AND TEST CROSSDIFFERENCE IN  BACK CROSS AND TEST CROSS
DIFFERENCE IN BACK CROSS AND TEST CROSS
 
Green chemistry and Sustainable development.pptx
Green chemistry  and Sustainable development.pptxGreen chemistry  and Sustainable development.pptx
Green chemistry and Sustainable development.pptx
 
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
 
GFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxGFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptx
 

Deep Learning Applications in Science

  • 1. Learning Systems for Science Ian Foster Argonne National Laboratory and The University of Chicago foster@anl.gov 1 Joint work with Rachana Ananthakrishnan, Ben Blaiszik, Kyle Chard, Ryan Chard, Mike Papka, Jim Pruyne, Steve Tuecke, Rick Wagner, Logan Ward, and others
  • 2. “Whatever you are studying right now if you are not getting up to speed on deep learning, neural networks, etc., you lose. We are going through the process where software will automate software, automation will automate automation.” -- Mark Cuban
  • 3.
  • 4. Deep leaning is also finding applications in science. Example: Predicting formation enthalpies of crystalline materials Best conventional machine learning method, Random Forest: a) Only elemental compositions (DFT-computed OQMD) Given DHf e.g.: Cr2Ni3 Al2O3 Predict: TiO2 ? Logan Ward et al., Phys Rev B, 2017
  • 5. Best conventional machine learning method, Random Forest: a) Only elemental compositions b) Also physical attributes (DFT-computed OQMD) Compute 145 physical properties: • Stoichiometric • Elemental property statistics • Electronic structure • Ionic compound Deep leaning is also finding applications in science. Example: Predicting formation enthalpies of crystalline materials Logan Ward et al., Phys Rev B, 2017
  • 6. Best conventional machine learning method, Random Forest: a) Only elemental compositions b) Also physical attributes Dipendra Jha ElemNet, 17-layer DNN: Only elemental compositions (Also runs 100x faster than RF.) 3,500550 Jha, Ward, et al., 2018. (DFT-computed OQMD) Deep leaning is also finding applications in science. Example: Predicting formation enthalpies of crystalline materials Logan Ward et al., Phys Rev B, 2017
  • 7. Deep learning • Drug response prediction • Scientific image classification • Scientific text understanding • Materials property design • Gravitational lens detection • Feature detection in 3D • Street scene analysis • Organism design • State space prediction • Persistent learning • Hyperspectral patterns Many other interesting applications are emerging Simulation • Materials science • Cosmology • Molecular dynamics • Nuclear reactor modeling • Combustion • Quantum computer simulation • Climate modeling • Power grid • Discrete event simulation • Fusion reactor simulation • Brain simulation • Transportation networks Big data • APS data analysis • HEP data analysis • LSST data analysis • SKA data analysis • Metagenome analysis • Battery design search • Graph analysis • Virtual compound library • Neuroscience data analysis • Genome pipelines Rick Stevens: Argonne applications for exascale
  • 8. Jack was right: It’s linear algebra all the way down https://xkcd.com/1838/
  • 9. We face many research challenges 9 Applications Learning systems Foundations Hardware Mathematics, algorithms; general AI, reinforcement learning, uncertainty quantification, explanability, etc. Advanced hardware to support AI. Evaluation of new architectures and systems. Neuromorphic and quantum as long-term AI accelerators? AI software. Software infrastructure for managing data, models, workflows etc., and for delivering AI capabilities to 10,000s of scientists and engineers. AI applications across science and engineering. New approaches to simulation and experimental science.
  • 10. DeepAI We need a lot more computing Exaflop/s-days used to train: AlexNet: 0.000007 (in 2012) AlphaGo Zero: 2 (in 2017) x 300,000 in 5.5 years
  • 11. Opportunities for science automation: Research today 11 Configure apparatus/write code Run experiments Solve societal problems Create knowledge What scientists want to do Most scientist time Analyze and plan
  • 12. 12 Run experiments Create knowledge Most scientist time AI assistants Analyze and plan Opportunities for science automation: Research tomorrow Solve societal problems Configure apparatus/write code
  • 13. Example: Accelerated discovery of metallic glasses Metallic glasses offer unique properties, but discovering new, useful alloys is slow • ML model predicts glass formation • Validate with automated experimentation • Active learning to optimize experiments 13 Ren et al. Sci Adv. (2017) eaaq1566
  • 14. Random forest to predict metallic glass formation Batch active learning to choose experiments Discovery of new ternary glass systems 14 Ren et al. Sci Adv. (2017) eaaq1566 Example: Accelerated discovery of metallic glasses
  • 15. 15 Imagine when only the fun parts of science remain https://twitter.com/worrydream/status/992546529217933312
  • 16. 16 Developing a DL model remains an artisanal process Model selection Model training Inference Training data Q A Training data Human expertise model architecture trained model
  • 17. Many challenges. For example … • Finding relevant models and methods (1000s of papers per year) • Finding relevant data for training and validation • Implementing, training, testing, and validating models • Configuring and adapting models • Scaling, accelerating, and optimizing models • Leveraging new architectures • Integrating models into scientific work processes • Documenting, sharing, and explaining results • Integrating and applying advanced methods: UQ, active learning, reinforcement learning, … • Engaging and educating the non-expert 99.99%
  • 18. 18 Learning systems AI software. Software infrastructure for managing data, models, workflows etc., and for delivering AI capabilities to 10,000s of scientists and engineers. “Without deep understanding of the basic tools needed to build and train new algorithms … researchers creating AIs resort to hearsay, like medieval alchemists. People gravitate around cargo-cult practices, relying on folklore and magic spells.” – Science, May 3 2018 New “learning systems for science”
  • 19. Organizing relevant data: Materials Data Facility EP EP EP • Query • Browse • Aggregate • Mint DOIs • Associate metadata • Persist datasets Databases Datasets APIs LIMS etc. Distributed data storage Data Publication Data Discovery materialsdatafacility.org Ben Blaiszik, Logan Ward, Jonathan Gaff, and others
  • 20. DLHub: A data and learning hub for science • Collect, publish, categorize models/code/ weights/data from many sources • Serve models via API to foster sharing, consumption, and access to data, training sets, and models • Automate training of models (using HPC as needed) as new data are available • Enable new science through reuse and synthesis of existing models TrainCollect Serve Ben Blaiszik, Ryan Chard, Logan Ward, and others
  • 21. “beam misaligned” “…” Say you want to use a deep neural network for online identification of problems when running diffraction experiments DLHub: Collect, serve, train community models
  • 23. ▪ Where are the model and trained weights? ▪ How do I run the model on my data? ▪ Should I run the model on my data? ▪ How can I retrain the model on new data? https://doi.org/10.1109/NYSDS.2017.8085045 DLHub: Collect, serve, train community models
  • 24. DLHub [“beam off image”, …] model/xray/batch_predict ▪ Where are the model and trained weights? ▪ How do I run the model on my data? ▪ Should I run the model on my data? ▪ How can I retrain the model on new data? https://doi.org/10.1109/NYSDS.2017.8085045 DLHub: Collect, serve, train community models
  • 25. DLHub [“beam off image”, …] model/xray/batch_predict ▪ Where are the model and trained weights? ▪ How do I run the model on my data? ▪ Should I run the model on my data? ▪ How can I retrain the model on new data? https://doi.org/10.1109/NYSDS.2017.8085045 DLHub: Collect, serve, train community models
  • 26. DLHub Collect Data 1) Register a model Train Model Register Model Model / transform containers Receive DOI Send to DLHub DLHub: Collect, serve, train community models
  • 27. DLHub Collect Data Receive predicted Properties Send compositions Call DLHub Find Model 2) Run a model Model / transform containers DLHub: Collect, serve, train community models Collect Data Receive DOI 1) Register a model Train Model Register Model Send to DLHub
  • 28. DLHub: Initial Use Cases • X-Ray diffraction (XRD) image tagging model • Prediction of bulk metallic glass forming regions in ternary diagrams • Predicting compound stability and bandgap by elemental composition Coming Soon • Deep learning to predict crystalline materials • ML/DL applied to high-throughput catalyst synthesis, simulation, and characterization • DL for chemical compound stability prediction • High-throughput High-Energy Diffraction Microscopy (HEDM) analysis with SLAC, NIST, NU, USC, Citrine With CHiMaD/NU Wang, Yager et al.
  • 30.
  • 31. 31
  • 33. Ben Blaiszik Steve TueckeKyle Chard Jim Pruyne Logan WardRachana Ananthakrishnan Ryan Chard Mike Papka Rick Wagner I reported on the work of many talented people Thanks also to: • Jon Almer, Francesco de Carlo, Hemant Sharma, Brian Toby, Stefan Vogt, Stephen Streiffer, Nicholas Schwarz, Doga Gursoy, and others, Advanced Photon Source • Tekin Bicer, Jonathan Gaff, Raj Kettimuthu, Justin Wozniak, and others, Argonne Computing We thank our sponsors DLHub Globus IMaD Petrel Argonne Leadership Computing Facility
  • 34. 34 Applications Learning systems Foundations Hardware Mathematics, algorithms; general AI, reinforcement learning, uncertainty quantification, explanability, etc. Advanced hardware to support AI. Evaluation of new architectures and systems. Neuromorphic and quantum as long-term AI accelerators? AI software. Software infrastructure for managing data, models, workflows etc., and for delivering AI capabilities to 10,000s of scientists and engineers. AI applications across science and engineering. New approaches to simulation and experimental science. “All the impressive achievements of deep learning amount to just curve fitting.” – Judea Pearl