Los Alamos National LaboratoryInformation Science and Technology Institute (ISTI)
Implementing and fostering collaborative research, workforce and program development, and technical exchange

2017 ISTI Data Challenge

Hosted by the Information Science and Technology Institute in collaboration with the Intelligence and Emerging Threats Program Office.

Contact  

  • Institute Director
  • Jim Ahrens
  • (505) 667-5797
  • Email
  • Professional Staff Assistant
  • Veronica Rodriguez
  • (505) 665-0528
  • Email

What are the goals of the challenge?

  • Discover new science and insights from challenging LANL data sets, including network and computer events from LANL’s enterprise network and physical sensor measurements from the Trinity supercomputer’s open science phase.
  • Showcase and test novel and established data science methods and tools.
  • Support training and professional development in data science through “prizes” of program development funds for participants.
  • Prepare the workforce for future efforts in data analytics.

What is the format?

  • Get an overview of the challenge data sets at the kickoff meeting April 6 where domain experts will introduce the kinds of questions they’d like to answer with their data.
  • Choose one or more challenge data sets to explore using methods you think will provide insights — e.g., visualization, machine learning, signal processing, statistics, or another strategy.
  • Work individually or as a team on questions you define for the challenge data sets you choose.
  • Present your results in a debrief session May 18, six weeks after the kickoff meeting, to be considered for one of the awards.

A panel of judges will evaluate the presentations and award two or more “prizes" of program development funds to be used toward training or professional development in data science in FY2017.

Who can participate?

Current LANL staff, postdocs, and students.

How can I access the challenge data sets and slides?

Take the following steps to gain access to the ISTI data challenge website:

Step 1: (for Foreign nationals only) Email your Z number to kary@lanl.gov and request the “CCS Collaboration” OCE role to access Darwin.
Step 2: Go to https://darwin.lanl.gov/ and log in with your moniker and cryptocard. If you don’t know your moniker, you can look it up at https://register.lanl.gov/findname/index.php​ (it’s in the “Owner UserID” field).
Step 3: Email kary@lanl.gov to say you’ve completed Step 1.
Step 4: You’ll receive an email when you’ve been added to the ISTI Data Challenge project so that you can access the data and forum as well as slides from the kickoff meeting and debrief session.

Questions? Contact Kary Myers, kary@lanl.gov.

Thursday, May 18, 1:30-5 p.m.

2017 ISTI Data Challenge Debrief

Hosted by the Information Science and Technology Institute (isti.lanl.gov) in collaboration with the Intelligence and Emerging Threats Program Office
Location: MSL Auditorium (TA-3, Building 1698, Room A103)

  • Welcome and intro, Kary Myers
  • Sunil Thulasidasan, CCS-3: LANSCE
  • Phil Romero, HPC-ENV (co-investigator Gopinath Chennupati, CCS-3): LANSCE, Trinity
  • Juston Moore, A-ARCS (co-investigator Lissa Baseman, HPC-DES): Trinity, Cyber #1 and #2
  • Shane Steinfadt, A-ARCS (co-investigators Daniel Byrne, Aaron Pope, A-ARCS): Cyber #1 and #3
  • Hugh Greenberg, HPC-ENV: Cyber #2
  • David Mascarenas, NSEC (co-investigator Yongchao Yang, NSEC): Cyber #2, Trinity
  • Judging panel discusses presentations and makes decisions to be announced Friday, May 19.

Challenge data sets:

  • LANSCE: a subset of operational data collected over a 19 day period at the Los Alamos Neutron Science Facility.
  • Trinity: physical sensor measurements from the Trinity supercomputer’s open science phase.
  • Cyber #1: 9 month time series of user-computer pairs
  • Cyber #2: 58 days of comprehensive, multi-source cybersecurity events
  • Cyber #3: ~3 months Windows event logs + network flow data