Name: Fast Machine Learning for Science Workshop 2022
Start: 2022-10-03T09:00:00-05:00
End: 2022-10-06T12:30:00-05:00
Location: Southern Methodist University

Fast Machine Learning for Science Workshop 2022

from Monday, 3 October 2022 (09:00) to Thursday, 6 October 2022 (12:30)

Monday, 3 October 2022
09:00 Welcome - Alessio Deiana (CERN) Allison Mccarn Deiana (Southern Methodist University (US))
Welcome
- Alessio Deiana (CERN)
- Allison Mccarn Deiana (Southern Methodist University (US))
09:00 - 09:05
09:05 Machine Learning for Science - Pan Li (Purdue University)
Machine Learning for Science
- Pan Li (Purdue University)
09:05 - 09:45
09:50 Efficient Computer Architectures - Sophia Shao (UC Berkeley)
Efficient Computer Architectures
- Sophia Shao (UC Berkeley)
09:50 - 10:30
10:30 Coffee
Coffee
10:30 - 11:00
11:00 Applications and Opportunities at the Large Hadron Collider - Maximilian J Swiatlowski (TRIUMF (CA))
Applications and Opportunities at the Large Hadron Collider
- Maximilian J Swiatlowski (TRIUMF (CA))
11:00 - 11:40
11:45 Application and Opportunities in Nuclear Physics - Jin Huang (Brookhaven National Lab)
Application and Opportunities in Nuclear Physics
- Jin Huang (Brookhaven National Lab)
11:45 - 12:25
12:30 Lunch
Lunch
12:30 - 13:30
13:45 Contributed Talks
Contributed Talks
13:45 - 15:15
Contributions

13:45 Designing intelligent DAQ systems for radiation instrumentation with hls4ml - Audrey Corbeil Therrien (Université de Sherbrooke)

14:00 Design and first test results of a reconfigurable autoencoder on an ASIC for data compression at the HL-LHC - Danny Noonan (Fermi National Accelerator Lab. (US))

14:05 Increasing the LHC Computational Power by integrating GPUs as a service - William Patrick Mccormack (Massachusetts Inst. of Technology (US)) Yongbin Feng (Fermi National Accelerator Lab. (US))

14:10 Exa.TrkX inference as-a-service - Yongbin Feng (Fermi National Accelerator Lab. (US))

14:15 Quantized ONNX (QONNX) - Jovan Mitrevski (Fermi National Accelerator Lab. (US))

14:30 Accelerating JEDI-net for jet tagging on FPGAs - Zhiqiang Que (Imperial College London)

14:45 Implementation of a pattern recognition neural network for live reconstruction using AI processors - Patrick Schwaebig

15:00 FKeras: A Fault Tolerance Library for DNNs - Olivia Weng
15:15 Coffee
Coffee
15:15 - 15:45
15:45 Contributed Talks
Contributed Talks
15:45 - 17:15
Contributions

15:45 Neural network accelerator for quantum control - Giuseppe Di Guglielmo (Fermilab)

16:00 End-to-End Vertex Finding for the CMS Level-1 Trigger - Benjamin Radburn-Smith (Imperial College (GB)) Christopher Edward Brown (Imperial College (GB))

16:15 Resource Efficient and Low Latency GNN-based Particle Tracking on FPGA - Bo-Cheng Lai Shi-Yu Huang

16:30 Neural Signal Compression System for a Seizure-Predicting Brain Implant in CMOS 28nm - William Lemaire

16:45 Exploring FPGA in-storage computing for Supernova Burst detection in LArTPCs - Benjamin Hawks (Fermi National Accelerator Lab)

17:00 Rapid Generation of Kilonova Light Curves Using Conditional Variational Autoencoder - Surojit Saha (Institute of Astronomy, National Tsing Hua University, Taiwan)
18:00

18:00 - 20:00
Tuesday, 4 October 2022
09:00 Applications and Opportunities for Machine Controls - Verena Kain (CERN)
Applications and Opportunities for Machine Controls
- Verena Kain (CERN)
09:00 - 09:40
09:45 Applications and Opportunities in Fusion Systems - Chris Hansen (University of Washington)
Applications and Opportunities in Fusion Systems
- Chris Hansen (University of Washington)
09:45 - 10:25
10:30 Coffee
Coffee
10:30 - 11:00
11:00 Autonomous experiments in scanning probe microscopy: opportunities for rapid inference and decision making - Rama Vasudevan (ORNL)
Autonomous experiments in scanning probe microscopy: opportunities for rapid inference and decision making
- Rama Vasudevan (ORNL)
11:00 - 11:30
11:30 Toward Real Time Decision-Making in Material Science Experiments - Mitra Taheri (JHU)
Toward Real Time Decision-Making in Material Science Experiments
- Mitra Taheri (JHU)
11:30 - 12:00
12:00 Some current challenges in materials measurements: an industrial perspective - Roger Proksch (Oxford Instruments)
Some current challenges in materials measurements: an industrial perspective
- Roger Proksch (Oxford Instruments)
12:00 - 12:30
12:30 Lunch
Lunch
12:30 - 13:30
13:30
13:30 - 15:15
Contributions

13:30 Demonstration of Machine Learning-assisted real-time noise regression in LIGO - Muhammed Saleem Cholayil (University of Minnesota)

13:45 Rapid Fitting of Band-Excitation Piezoresponse Force Microscopy Using Physics Constrained Unsupervised Neural Networks - Alibek Kaliyev (Lehigh University)

14:00 Data Driven Weather Forecasting with Rudimentary Observables - Luke Fairbanks (ucsd)

14:05 Low-latency Noise Subtraction of Gravitational Wave Data by DeepClean - Chia-Jui Chou (National Yang Ming Chiao Tung University)

14:10 Harnessing ultrafast ML for new algorithms at the CMS L1 trigger - Daniel Diaz (Univ. of California San Diego (US))

14:15 Intelligent experiments through real-time AI: Fast Data Processing and Autonomous Detector Control for sPHENIX and future EIC detectors - Micol Rigatti (Fermi National Accelerator Lab. (US))

14:30 Application of deep learning to instability tracking using high-speed video cameras in magnetic confinement fusion - Yumou Wei (Columbia University)

14:45 Fast recurrent neural networks on FPGAs with hls4ml - Elham E Khoda (University of Washington (US))

15:00 Deep Neural Network Algorithms in the CMS Level-1 Trigger - Anthony Vizcaino Aportela (Univ. of California San Diego (US))
15:15 Coffee
Coffee
15:15 - 15:45
15:45
15:45 - 17:15
Contributions

15:45 Extremely Noisy 4D-TEM Strain Mapping Using Cycle Consistent Spatial Transforming Autoencoders - Shuyu Qin

16:00 Quantized Distilled Autoencoder Model on FPGA for Real-Time Crystal Structure Detection in 4D Scanning Transmission Electron Microscopy - Ryan Forelli (Lehigh University)

16:15 Deployment of ML in changing environments - Benjamin Radburn-Smith (Imperial College (GB)) Christopher Edward Brown (Imperial College (GB)) Marco Barbone

16:30 Low-latency Calorimetry Clustering at the LHC with SPVCNN - Alexander Joseph Schuy (University of Washington (US))

16:45 CryoAI – Prototyping cryogenic chips for machine learning at 22nm - Manuel Valentin (Northwestern University)

17:00 Next Generation Coprocessors as a service - Dylan Sheldon Rankin (Massachusetts Inst. of Technology (US))
18:00

18:00 - 21:00
Wednesday, 5 October 2022
09:00 Beyond CMOS - Jennifer Hasler (Georgia Tech)
Beyond CMOS
- Jennifer Hasler (Georgia Tech)
09:00 - 09:40
09:45 Applications and Opportunities in Neutrino and DM experiments - Jonathan Asaadi (University of Texas at Arlington) Jonathan Asaadi (University of Texas at Arlington (US))
Applications and Opportunities in Neutrino and DM experiments
- Jonathan Asaadi (University of Texas at Arlington)
- Jonathan Asaadi (University of Texas at Arlington (US))
09:45 - 10:25
10:30 Coffee
Coffee
10:30 - 11:00
11:00 Benchmarking Fast ML systems - Vijay Reddi (Harvard University)
Benchmarking Fast ML systems
- Vijay Reddi (Harvard University)
11:00 - 11:40
11:45 The need for low latency with billion-year old signals - Matthew Graham (California Institute of Technology) Matthew Graham (Caltech)
The need for low latency with billion-year old signals
- Matthew Graham (California Institute of Technology)
- Matthew Graham (Caltech)
11:45 - 12:25
12:30 Lunch
Lunch
12:30 - 13:30
13:30
13:30 - 15:15
Contributions

13:30 A Deep Learning Approach to Particle Identification for the AMS Electromagnetic Calorimeter - Raheem Hashmani (Middle East Technical University (TR))

13:45 A Normalized Autoencoder for LHC triggers - Luigi Favaro

14:00 Interaction Network Autoencoder in the Level-1 Trigger - Sukanya Krishna (Univ. of California San Diego (US))

14:05 Large CNN for HLS4ML and Deepcalo - ChiJui Chen Lin-Chi Yang Yan-Lun Huang

14:10 End-to-end acceleration of machine learning in gravitational wave physics - Alec Gunny (Massachusetts Institute of Technology)

14:15 FastML Science Benchmarks: Accelerating Real-Time Scientific Edge Machine Learning - Jules Muhizi (Fermilab/Harvard University)

14:30 Robust anomaly detection using NuRD - Abhijith Gandrakota (Fermi National Accelerator Lab. (US))

14:45 Quantized Neural Networks on FPGAs using HAWQ-V3 and hl4ml - Javier Ignacio Campos (Fermi National Accelerator Lab. (US))

15:00 Implementing Deep Neural Network Algorithms inside the CMS Level-1 Trigger - Duc Hoang (MIT)
15:15 Coffee
Coffee
15:15 - 15:45
15:45
15:45 - 17:15
Contributions

15:45 Real-time image processing for high-resolution imaging detectors - Georgia Karagiorgi

16:00 Semi-supervised Graph Neural Networks for Pileup Noise Removal - Shikun Liu

16:05 A Machine Learning Software Infrastructure for Gravitational Wave Signal Discovery - Ethan Marx (MIT)

16:10 Online-compatible Unsupervised Non-resonant Anomaly Detection - Vinicius Massami Mikuni (Lawrence Berkeley National Lab. (US))

16:15 In-System Parameter Update and I/O Capture for Machine Learning IP Cores - Brett McMillian (Crossfield Technology LLC)

16:30 Detection for Core-collapse Supernova and Fast Data Preprocessing - Andy Chen (National Yang Ming Chiao Tung University (NYCU))

16:35 EMD Neural Network Loss for ECON-T ASIC Autoencoder - Rohan Shenoy (Univ. of California San Diego (US))

16:40 A novel ML-based method of primary vertex reconstruction in high pile-up condition - Haoran Zhao (University of Washington (US))

16:45 In-Pixel AI: From Algorithm to Accelerator - Priyanka Dilip (Stanford University / Fermilab)

17:00 Autonomous real-time science-driven follow-up of survey transients - Niharika Sravan
Thursday, 6 October 2022
09:00 hls4ml tutorial - Sioni Paris Summers (CERN) Elham E Khoda (University of Washington (US))
hls4ml tutorial
- Sioni Paris Summers (CERN)
- Elham E Khoda (University of Washington (US))
09:00 - 11:00
11:00 hls4ml community session
hls4ml community session
11:00 - 12:30