Extreme Physics, Extreme Data
13 - 17 January 2020
Venue: Lorentz Center@Oort
If you are invited or already registered for this workshop, you have received login details by email.The aim of this workshop is to bring plasma physicists, astrophysicists and data scientists together to work on understanding the emerging statistical and data challenges in understanding extreme high energy density physics. The field is in the process of entering the regime of being data dominated in the way that astronomy has over the last ten years, and particle physics many years before that. The advent of high shot-rate, high power laser facilities, as well as the capacity to do hundreds of millions of simulations with exascale computing is leading to a need to understand uncertainty, and to leverage machine learning and modern algorithms to get the most out of our data in the coming decade.
Fundamental Problems:
INFERENCE: How do we accurately quantify uncertainties on physical inferences from data? SYNTHESIS: How do we combine data from different sources, or extrapolate results from one source to another e.g. to future larger facilities, or to astrophysical scale plasmas? DESIGN: How can we use algorithms to design experiments?
The workshop will focus on the application of modern statistical techniques, machine learning and algorithmic approaches to experiments and astronomical observations around the world probing the physics of the highest temperatures and pressures. This is a research area that lies at the intersection of astrophysics, plasma physics, and particle physics - researchers in these areas all have a shared interest in getting the most information out of their HEDP data.
Monday January 13
09:30-10:30 Registration
10:30-10:45 Welcome by Lorentz Center
10:45-11:00 P. Hatfield Talk Inference, Synthesis & Design Session Chair: C. Palmer
11:00-11:30 D. Mariscal Talk Problems in HEDP
11:30-12:00 G. Lapenta Talk Space Weather Prediction and Exascale Computing
12:00-13:30 Lunch@ Snellius restaurant
Key Statistical Methods
Session Chair: J. Kirschner
13:30-13:50 J. Gaffney Tutorial Talk Machine Learning and Bayesian Statistics
13:50-14:10 M. Kasim Tutorial Talk Optimization Methods
14:10-15:00 Lightning Talks
15:00-15:30 Coffee and tea break
15:30-16:30 G. Anderson Discussion What do we want to get out of the week?
16:30-17:00 D. Mariscal Summarize / Drafting
17:00- Wine and cheese welcome party
Tuesday January 14
Inference
Session Chair: L. Antonelli
09:00-09:30 L. Roso Talk Science at the Center for Pulsed Lasers with High Repetition Rate Lasers
09:30-10:00 T. Nagayama Talk Statistical analysis of Z stellar opacity data
10:00-10:30 Coffee and tea break
10:30-11:00 B. Kettle Talk Turning Light into Matter
11:00-11:30 C. Shneider Talk Machine Learning in Heliophysics
11:30-12:00 J. Citrin Talk Reduced Models for Tokamak Turbulence
12:00-13:30 Lunch@ Snellius restaurant
Key Issues in Uncertainty Quantification
Session Chair: K. Bennet
13:30-14:00 M. Kasim Talk Inverse Problem Instabilities
14:00-14:30 G. Anderson Talk UQ in Deep Learning
14:30-15:00 M. Fajardo Talk
15:00-15:30 Coffee and tea break
15:30-16:30 Parallel Sessions
16:30-17:00 S. Mangles Drafting Summary of UQ Methods
Wednesday January 15
Experiment Design
Session Chair: G. Sarri
09:00-09:30 V. Gopalaswamy Talk Tripling Yield with Algorithms
09:30-10:00 M. Martin Talk Automatic Design of Opacity Experiments
10:00-10:30 Coffee and tea break
10:30-11:00 M. Migliorini Talk Machine Learning Pipelines with Modern Big Data Tools for High Energy Physics
11:00-11:30 P. Hatfield Talk Automatic Design of ICF Experiments
11:30-12:00 M. Streeter Talk Active feedback laser pulse shaping
12:00-13:30 Lunch@ Snellius restaurant
New Algorithms and Computing for Design
Session Chair: T. Goffrey
13:30-14:00 J. Kirschner Talk SwissFEL and Bayesian Optimisation
14:00-14:30 L. Peterson Talk Next-gen algorithms & HPC
14:30-15:00 Lightning Talks
15:00-15:30 Coffee and tea break
15:30-16:30 S. Rose Discussion Optimal Use of Exascale Computing
16:00-17:00 Summarize / Drafting
17:00-17:30 Departure by bus to workshop dinner
17:30-21:30 Workshop dinner, Boat on Kaag Lakes
21:30-22:00 Bus back to Central Station, Lorentz Center and Hotel
Thursday January 16
Synthesis of Models and Data
Session Chair: P. Knapp
09:00-09:30 K. Humbird Talk Transfer Learning in ICF
09:30-10:00 J. Gaffney Talk Model Calibration in ICF
10:00-10:30 Coffee and tea break
10:30-11:00 M. Macdonald Talk Multiple Diagnostics in HEDP experiments
11:00-11:30 S. Basegmez Du Pree Talk Cosmic data mining in Space
11:30-12:00 C. Palmer Talk Lab Astro with 10Hz Lasers
12:00-13:30 Lunch@ Snellius restaurant
Synthesis Challenges
Session Chair: M. Martin
13:30-14:00 B. Kustowski Talk Correcting Simulation Bias in Multi-Modal Data Using Transfer Learning
14:00-14:30 V. Hafych Talk Application of Parallel MCMC Sampling for Proton Bunch Analysis in the AWAKE Experiment
14:30-15:00 Lightning Talks
15:00-15:30 Coffee and tea break
15:30-16:30 Parallel Sessions
16:30-17:00 J. Ruby Summarize / Drafting
Friday January 17
Conclusions
Session Chair: C. Shneider
09:00-09:30 S. Ali Talk Using experimental uncertainties to build uncertainty-aware equations of state for extreme conditions
09:30-10:30 J. Gaffney Wrap Up Discussion
10:30-11:00 Coffee and tea break
11:00-12:20 Drafting
12:20-12:30 P.Hatfield Talk Concluding Remarks
12:30 Lunch@ Snellius restaurant
