July 17th, 2023
| Title | Speaker | Time | Duration |
|---|
| DAY 1, SESSION 1 – (Topic: Large Language Models / LLMs & NLP) |
| Opening Remarks | Tyson Bailey (SNL) | 9:00am | 10 minutes |
| Keynote #1: Panel discussion on LLMs | Scott Steinmetz (SNL) | 9:10am | 60 minutes |
| Q&A with LLM panel | Scott Steinmetz (SNL) | 10:10am | 20 minutes |
| Advancements in Machine Learning | Christopher Symonds (SNL) | 10:30am | 30 minutes |
| – 5 MIN BREAK – |
| Unlocking the Power of Large Language Models: Practical Applications in Scripts and Programs | Anthony Garland (SNL) | 11:05am | 20 minutes |
| Searching for Meaning: Transfer Learning with LLMs to Improve Search Results | Sarah Ackerman (SNL) | 11:25am | 20 minutes |
| The Cost of Custom Large Language Model (LLM) Solutions | Henry Wong (SNL) | 11:45am | 20 minutes |
| The Potential Dangers of Generative AI via Reification of Historical Conflict Strategies | Ruby E. Booth (SNL) | 12:05pm | 30 minutes |
| – LUNCH BREAK (12:05pm – 1pm) – |
| AI-enhanced simulation of operational requirements management with a knowledge graph-based digital twin | Matthew T. Dearing (ANL) | 1:00pm | 20 minutes |
| ASPER: ASP-enhanced Deep Learning Models for Joint Entity-Relation Extraction | Trung Le (New Mexico State University) | 1:20pm | 15 minutes |
| Questions & Answers for Day 1, Session 1 | | 1:35pm | 15 minutes |
| – 10 MIN BREAK – |
DAY 1, SESSION 2 (Multiple Tracks:
RECRUITING SESSION can be found below, CUI SESSION see internal schedule)
|
| Opening Remarks | Elihu Deneke (SNL) | 2:00 PM | 5 minutes |
| Idaho National Laboratory | TBD | 2:05 PM | 20 minutes |
| Los Alamos National Laboratory | TBD | 2:25 PM | 20 minutes |
| Oak Ridge National Laboratory | TBD | 2:45 PM | 20 minutes |
| Sandia National Laboratories | TBD | 3:05 PM | 20 minutes |
| Questions | | 3:25 PM | 5 minutes |
MIXER: Nuclear Museum at 6:00 pm
July 18th, 2023
| Title | Speaker | Time | Duration |
|---|
| DAY 2, SESSION 1 – (Topic: Climate Science) |
| Opening Remarks | Tyson Bailey (SNL) | 9:00am | 5 minutes |
| Keynote #2: Selected Machine Learning Applications in Weather and Climate | Paul Roebber (University of Wisconsin) | 9:05am | 30 minutes |
| Block Dynamic Mode Decomposition: Domain-Aware Machine Learning for Climate Modelling and Feedback Analysis | Craig Bakker (PNNL) | 9:35am | 15 minutes |
| Surrogate model coupling in the Energy Exascale Earth System Model | Kevin Yarritu (SNL) | 9:50am | 20 minutes |
| – 5 MIN BREAK – |
| Characterizing climate pathways using feature importance on echo state networks | Katherine Goode (SNL) | 10:15am | 20 minutes |
| Solving High-Dimensional Inverse Problems with Auxiliary Uncertainty via Operator Learning with Limited Data | Mamikon Gulian (SNL) | 10:35am | 20 minutes |
| Questions & Answers for Day 2, Session 1 | | 10:55am | 15 minutes |
| – LUNCH BREAK (11:10am – 12:00pm) – |
| DAY 2, SESSION 2 – (Topic: Physics-Informed Machine Learning) |
| Data-Driven Structure Preservation for Scientific Machine Learning | Jonas Actor (SNL) | 12:00pm | 20 minutes |
| Spatio-Temporal Super-Resolution of Dynamical Systems using Physics-Informed Deep-Learning | Ankit Shrivastava (SNL) | 12:20pm | 15 minutes |
| Manifold Learning & Equations of State: Conservation of Energy as Regularization for Neural Networks | George A. Kevrekidis (LANL) | 12:35pm | 20 minutes |
| – 5 MIN BREAK – |
| Multiscale Damage via Physics-Informed Recurrent Neural Networks | Ramin Bostanabad (UC Irvine) | 1:00pm | 20 minutes |
| Utilizing Physics-informed and Machine Learning Methods to Enhance Remote Monitoring of Physiological Signatures | Christopher Katinas (SNL) | 1:20pm | 20 minutes |
| Closure modeling through the lens of multifidelity operator learning | Shady Ahmed (PNNL) | 1:40pm | 15 minutes |
| Questions & Answers for Day 2, Session 2 | | 1:55pm | 15 minutes |
| – 5 MIN BREAK – |
| DAY 2, SESSION 3 – (Topic: Nuclear Science) |
| Machine Learning Surrogates for Time Dependent Fuel Degradation Processes in Nuclear Waste Repository Simulations | Bert Debusschere (SNL) | 2:15pm | 20 minutes |
| Transferable and explainable reactor power level models using nonradiological data | Jon Whetzel (SNL) | 2:35pm | 15 minutes |
| A Semi-Supervised Model for Multi-Label Radioisotope Classification and Out-of-Distribution Detection | Alan Van Omen (SNL) | 2:50pm | 30 minutes |
| Questions & Answers for Day 2, Session 3 | | 3:20pm | 15 minutes |
| MIXER: NONE |
July 19th, 2023
| Title | Speaker | Time | Duration |
|---|
| DAY 3, SESSION 1 – (Topic: All Things Data) |
| Opening Remarks | Tyson Bailey (SNL) | 9:00am | 5 minutes |
| Keynote #3: Want To Do AI? Bring a Large Toolbox! | George Luger (UNM) | 9:05am | 30 minutes |
| Battery Cycle Life Prediction Across a Variety of Datasets | Joseph Lubars (SNL) | 9:35am | 15 minutes |
| Autoencoders for Multimodal Data Fusion | Georgios Fragkos (SNL) | 9:50am | 15 minutes |
| Leveraging 3D Game Engines for High-Quality Synthetic Data in Object Detection Applications | Lucas Zhou (SNL) | 10:05am | 15 minutes |
| – 5 MIN BREAK – |
| Synthetic data generation to improve object detection in x-ray radiographs | Heidi Komkov (SNL) | 10:25am | 15 minutes |
| Quantifying the Sensitivity of Magnetohydrodynamics Simulations of Pulsed-Power Experiments Using Multi-Fidelity Electrical Conductivity Data | Lucas Stanek (SNL) | 10:40am | 20 minutes |
| Turtle: A Tool for Catching Subpopulation Trends in Performance Data | Skyler Gray (SNL) | 11:00am | 15 minutes |
| Probabilistic Neural Data Fusion for Learning from an Arbitrary Number of Multi-fidelity Data Sets | Carlos Mora Sardiña (UC Irvine) | 11:15am | 20 minutes |
| Questions & Answers for Day 3, Session 1 | | 11:35am | 15 minutes |
| – LUNCH BREAK (11:50am – 1:00pm) – |
| DAY 3, SESSION 2 – (Topic: Material Science & Manufacturing) |
| Multi-Fidelity Bayesian Optimization for Materials Design | Zahra Zanjani Foumani (UC Irvine) | 1:00pm | 20 minutes |
| PRISM: Process Parameter Optimization for Selective Manufacturing | Anthony Garland (SNL) | 1:20pm | 15 minutes |
| Adaptive spatiotemporal dimension reduction in concurrent multiscale damage analysis | Shiguang Deng (UC Irvine) | 1:35pm | 15 minutes |
| – 10 MIN BREAK – |
| Deep Reinforcement Learning for the Design of Mechanical Metamaterials with Customizable Nonlinear Deformation Responses | Nathan Brown (SNL) | 2:00pm | 20 minutes |
| Hydrodynamic parameter estimation using statistical machine learning for dynamic radiography | Soumi De (LANL) | 2:20pm | 20 minutes |
| Artificial Intelligence and Machine Learning Support for Probabilistic Fracture Mechanics Analysis | Stephen Verzi (SNL) | 2:40pm | 20 minutes |
| Multivariate Analysis as a Tool for Validating Tester Matching | Rosalie Multari (SNL) | 3:00pm | 20 minutes |
| Questions & Answers for Day 3, Session 2 | | 3:20pm | 15 minutes |
| MIXER: Apothecary Rooftop Lounge at 6pm |
July 20th, 2023
| Title | Speaker | Time | Duration |
|---|
| DAY 4, SESSION 1 – (Topic: Theory & Applications) |
| Opening Remarks | Tyson Bailey (SNL) | 9:00am | 5 minutes |
| Keynote #4: Deep learning estimation of modified Zernike coefficients for image point spread functions | Steven Sandoval (New Mexico State University) | 9:05am | 30 minutes |
| Trusted AI to Aid HPC Software Development for U.S. DoE’s Scientific Software | Vivek Kale (SNL) | 9:35am | 20 minutes |
| Machine Learning and Autonomous Decision Making for National Security: Designing Decision Rules using Multiobjective Optimization | Mark A. Smith (SNL) | 9:55am | 20 minutes |
| Towards High-Speed Forecasting of Destabilizing Events: a Ukrainian Case Study | Chrysm Watson Ross (SNL) | 10:15am | 15 minutes |
| – 5 MIN BREAK – |
| Real-time decision-making support for risk-reduction during severe accidents | Alex Washburne (SNL) | 10:35am | 15 minutes |
| AI-enabled conflict simulation | Ephraim Rusu (LLNL) | 10:50am | 15 minutes |
| Bayesian forecasting of small subgraph events in temporal networks | Matt Hoffman (SNL) | 11:05am | 20 minutes |
| Learning Wildfire Dynamics for Ignition Inversion | Zachary Morrow (SNL) | 11:25am | 15 minutes |
| Questions & Answers for Day 4, Session 1 | | 11:40am | 15 minutes |
| – LUNCH BREAK (11:55am – 1pm) – |
| DAY 4, SESSION 2 – (Topic: Theory & Applications) |
| Applying Machine Learning in the Characterization of Ship Tracks | Pierce Warburton (SNL) | 1:00pm | 15 minutes |
| Behavioral Segmentation and Clustering of Geospatial Trajectories | Jessica Jones (SNL) | 1:15pm | 20 minutes |
| Tracking Walking Droplets | Erdi Kara (Spelman College) | 1:35pm | 20 minutes |
| Statistical modeling for detecting anomalous event timings on host-based cybersecurity log data | Alexander Foss (SNL) | 1:55pm | 20 minutes |
| – 5 MIN BREAK – |
| Deep Offline Reinforcement Learning with Recurrent Training for Cyberphysical Threat Detection | Georgios Fragkos (SNL) | 2:20pm | 20 minutes |
| Sensor and Actuator Attacks on Hierarchical Control Systems with Domain-Aware Operator Theory | Craig Bakker (PNNL) | 2:40pm | 15 minutes |
| Poisson Response Tensor-on-Tensor Regression | Carlos Llosa (SNL) | 2:55pm | 20 minutes |
| Class-Specific Attention (CSA) for Classifying Time-Series Data | Huiping Cao (New Mexico State University) | 3:15pm | 20 minutes |
| Questions & Answers for Day 4, Session 2 | | 3:35pm | 15 minutes |
| Closing Remarks | Tyson Bailey (SNL) | 3:50pm | 10 minutes |
| MIXER: Marble Brewery at 6pm |