CLSAC
  • Home
  • Events
    • CLSAC'12
    • CLSAC'13
    • CLSAC'14
    • CLSAC'15
    • CLSAC'16
    • CLSAC'17
    • CLSAC'18
    • CLSAC'19
    • CLSAC'20
    • CLSAC'21
    • CLSAC'22
    • CLSAC'23
    • CLSAC'24
    • CLSAC'25
  • George Cotter Award
  • Contact
Picture
Photo credit: Westin Annapolis, Annapolis Maryland

CLSAC 2018

High Impact Large-Scale Analytics

October 30--November 1, 2018
The Westin Annapolis
100 Westgate Circle
Annapolis, MD  21401

Theme: 
​The seventh CLSAC workshop will examine the theory and practice of high consequence analytics at scale in industry, science, defense, and law enforcement.  Using predictions, classification, discovered relationships, and decision sets, high consequence analytics create actionable insights that impact human life, critical infrastructure, national economies, international relationships, and social institutions.  This year’s focus will examine the unique nature of high consequence analytics at scale from the perspective of taking action, understanding how the proposed action was derived, the critical sources of data, and the unique systems-level challenges of building High Consequence systems.   These systems constitute unique software and hardware environments capable of handling large data sets with hard predictive deadlines. They must be credible, trustworthy, scalable, resilient, and resistant to attack.  The results must reflect high confidence.

CLSAC brings together thought leaders across industry, government, and academia to discuss critical challenges for high consequence analytics.  Important questions include: 
  • What motivates the community to find solutions (e.g., what are the consequences)?
  • What requirements are imposed by the scale of the data and the variety of data sources available?
  • What are the privacy and other ethical requirements of using such a system?
  • How do we determine whether or not action should be taken?
  • How do we audit such action?
  • What do the hardware and software systems need to do to achieve these goals?
  • What are the human factors?
Organizing Committee:
John Feo (PNNL)
​Jim Ang (PNNL)

David Haglin (Trovares Inc.)
Richard Murphy (Micron Technology)
Ron Oldfield (SNL)
Steve Pritchard (DoD)
Tyler Simon (DoD)
​T.C. Tuan (DoD)
Candace Culhane (LANL)
         2018 Sponsors
Picture
Picture
Picture
Picture
Picture
Picture
Picture
Picture
Picture
Picture
Picture
Picture
Picture

Agenda


2018_clsac_agenda.pdf
File Size: 368 kb
File Type: pdf
Download File

Tuesday, October 30

7:00 - 8:30 am
Breakfast and Registration (Capitol ABC)
8:00 - 8:30 am
Welcome
​George Cotter Award for Vision and 
Leadership in the Field of Data Analytics
John Feo, PNNL
Richard Murphy, Micron
Candy Culhane, LANL
Keynote
8:30 - 9:15 am
Machine Learning to Predict Cancer Drug Response:  Problems in Model Transfer and Uncertainty Quantification
Rick Stevens, ANL​
Session 1: Applications 1
Tyler Simon, Moderator
9:15 - 10:00 am
Petabyte-Scale Cyber Security Analytics to Discover and Investigate Targeted Attacks
Alejandro Borgia, Symantec
10:00 - 10:15 am
Break
10:15 - 11:00 am
Detecting Lateral Movement with a Compute-intense Graph Kernel
Steve Reinhardt, D-Wave Systems
11:00 - 11:45 am
Geospatial Cloud Analytics
Dolores Shaffer, Science & Technology Associates
11:45 - 12:30 pm
Panel Discussion
12:30-1:45 pm
Lunch (Capitol ABC)
Session 2: Software Enablers
Ron Oldfield, Moderator
1:45 - 2:30 pm
High-Performance Portable Data Analytics Software Using Kokkos
Michael Wolf, SNL
2:30 - 3:15 pm
How I Learned to Stop Worrying About BGL and Learned to Love the C++ (11) Standard Library
Andrew Lumsdaine, PNNL
3:15 - 3:30 pm
Break
3:30 - 4:15 pm
ENSIGN Advanced Tensor Decompositions to Find Explanations
Richard A. Lethin, Reservoir Labs
4:15 - 5:00 pm
Panel Discussion
6:00 - 7:30 pm
Dinner (Capitol ABC)
The Ultimate Baseball Roadtrip and Algorithm
Michael Mountain, Catalyte
7:30 - 9:00 pm
Reception (Senate AB)

Wednesday, October 31

7:30 - 8:30 am
Breakfast (Capitol ABC)
Keynote
8:30-9:15am
Engineering Automated Analytics Systems
Robert Bonneau, DoD
Session 3: Applications 2
David Haglin, Moderator
9:15 - 10:00 am
Back to basics: applying classic methods to the design of machine learning hardware
Doug Wightman, GROQ
10:00-10:15 am
Break (Capitol ABC Foyer)
10:15 - 11:00 am
Developing Tractable Adaptive Data Collection and Archiving Plans for Large-scale Cyber Networks​
Georgiy Levchuk, Aptima
11:00 - 11:45 am
​Quantifying Topological Uncertainty in Fractured Systems Using Graph Theory and Machine Learning
​Gowri Srinivasan, LANL
11:45 - 12:30 pm
Panel discussion
12:30-6:00 pm
Free Time (lunch and dinner on your own)
6:00-7:30 pm
Random Access Session (Capitol Ballroom)
Sign-up sheet will be at the registration table.
​Talks limited to 10 minutes. 
John Feo, Moderator
7:30-9:00 pm
Reception and Poster Session (Senate AB)
Poster
Student
​HPC Job Outcome Prediction: System Log Feature Extraction and Importance
​Alexandra DeLucia, 
Los Alamos National Lab
​Optimizations for Sparse Matrix-Vector Multiply on a Migratory Thread Architecture
​Thomas Rolinger, University of Maryland
​Distributed Memory Graph Algorithms: Case Studies with Community Detection and Weighted Matching
​Sayan Ghosh, Washington State University at Pullman
Fast Linear Algebra-Based Triangle Analytics with Kokkos Kernels
​Abdurrahaman Yasar
Georgia Tech

Thursday, November 1

7:00 - 8:30 am
Breakfast
Keynote
8:30 - 9:15 am
The Rise of Learning Machines
Rich Friedrich, DoD
Session 4: Hardware Enablers
Candy Culhane, Moderator
9:15 - 10:00 am
​Future memory architectures for large-scale analytics
Anton Korzh, Micron, Inc.
10:00 - 10:15 am
Break (Capitol ABC Foyer)
10:15 - 11:00 am
DGX-2 Graph Analysis Performance for High- Consequence Analytics​
Brad Rees/Joe Eaton, NVIDIA​
11:00 - 11:45 pm
Modern Software Approaches for Operationalizing the Application of Machine Learning/AI
Leonard Walstad, Pivotal
11:45 - 12:30 pm
Panel Discussion
12:30 - 1:45 pm
Lunch (Capital ABC)
Session 5: Methods and Environments
Jim Ang, Moderator
1:45 - 2:30 pm
​Toward Socio-Cultural Machine Learning
Mark Riedl, GA Tech
2:30 - 3:15 pm
Evolving Better AI's by Leveraging Leadership Class Supercomputing
Travis Johnston, ORNL
3:15 - 3:30 pm
Break (Capitol ABC Foyer)
3:30 - 4:15 pm
Classification without Representation: Interactive Machine Learning at Scale with CHISSL
Dustin Arendt, PNNL
4:15 - 5:00 pm
Panel Discussion
5:00 - 5:15 pm
Closing Remarks and Adjourn
5:15 - 7:00 pm
Reception (Senate A/B)
The CLSAC web site is sponsored by the Association for High-Speed Computing.