Name: NGT - Openlab "Optimising Floating Point Precision" Workshop
Start: 2025-07-01T13:00:00+02:00
End: 2025-07-02T18:00:00+02:00
Location: CERN

NGT - Openlab "Optimising Floating Point Precision" Workshop

from Tuesday 1 July 2025 (13:00) to Wednesday 2 July 2025 (18:00)
CERN (40/S2-B01)

: Sessions

: Talks

: Breaks

1 Jul 2025

2 Jul 2025

09:00	Using physics knowledge to improve numerical stability - Enrico Bothmann (CERN) (40/S2-B01 - Salle Bohr) bothmann-numerical-stability-2025-07-02.pdf Enrico_Bothmann.mp4
09:30	Double-double for virtual amplitude evaluation - Vitaly Magerya (CERN) (40/S2-B01 - Salle Bohr) magerya-dd.pdf Vitaly_Magerya.mp4
10:00	An overview of mixed precision strategies for scientific computing - Theo Mary (Computer Lab of Paris 6 (Lip6)) (40/S2-B01 - Salle Bohr) main_CERN.pdf Theo_Mary.mp4
10:30	Adaptive Floating-Point Quantization for Efficient Neural Networks - Nicolo Ghielmetti (CERN) float-workshop-openlab.pdf Nicolo_Ghielmetti.mp4
10:50	Discussion - Questions (40/S2-B01 - Salle Bohr)
11:05	--- Coffee ---
11:35	Mixed precision ab initio tensor network state methods adapted for NVIDIA. Blackwell technology via emulated FP64 arithmetic - Ors Legeza (Wigner Research Centre for Physics, Hungary) (40/S2-B01 - Salle Bohr) cern2025_v01.pdf Ors_Legeza.mp4

14:00	Welcome Session - Stefan Roiser (CERN) 20250601-WS-Float.pdf
14:10	Problem statement from experiments and SFT - Vincenzo Innocente (CERN) Arith@CERN2025.pdf
14:40	Problem statement from Theory Department - Jacob Finkenrath HFloat_at_TH1.pdf
14:50	Discussion - Questions
15:00	--- Coffee ---
15:30	Floating Point Emulation in NVDIA Math Libraries - Samuel Rodriguez (NVidia) cern-talk.pdf Samuel_Rodriguez.mp4
16:15	Advancing AI with AMD: Open Source, Sovereign Innovation, and the Latest in CPU & GPU Performance - Sayed Maudodi Joerg Roskowetz AMD_CERN_workshop_July_1st_2025.pdf Joerg_Roskowetz.mp4
16:45	VXP: Extended Precision Accelerator - Eric Guthmuller (CEA France) 20250701_CERN_workshop_EG.pdf 20250701_CERN_workshop_EG.pptx Eric_Guthmuller.mp4
17:05	Extended Precision in Convex Optimisation - David Herrera-Marti (CEA France) 20250701_CERN_workshop.pdf David_Herrera_Marti.mp4
17:25	Discussion - Questions
17:45	--- Networking Cocktail ---

12:05	TNL: Numerical Library for Modern Parallel Architectures - Thomas Oberhuber (Czech Technical University in Prague) (40/S2-B01 - Salle Bohr) cern-tnl.pdf Thomas_Oberhuber.mp4 video_ikem_qcrit_normal_flux.mp4
12:20	Float32 Expansions – A Possible Answer for Scientific Computing in the Era of AI-Driven GPU Development - František Stloukal (Czech Technical University in Prague) Frantisek_Stloukal.mp4 Optimising_floating_point_precision___CERN___TNL-4.pdf
12:35	Discussion - Questions (40/S2-B01 - Salle Bohr)
13:05	--- Lunch ---
14:20	Floating-Point Error Estimation Using Automatic Differentiation - Vassil Vasilev (Princeton University) (40/S2-B01 - Salle Bohr) ChefFP.pdf Vassil_Vasilev.mp4
14:50	Precision auto-tuning and control of accuracy in high performance simulations - Fabienne Jézéquel (LIP6, Sorbonne Université) (40/S2-B01 - Salle Bohr) Fabienne_Jezequel.mp4 slides_FJezequel.pdf
15:20	Experiences with CADNA and the Madgraph5 Event Generator - Stephan Hageboeck (CERN) (40/S2-B01 - Salle Bohr) CADNA Madgraph.pdf Stephan_Hageboeck.mp4
15:50	--- Coffee ---
16:20	Emulating Matrix Multiplication Using Mixed-Precision Computation - Katsuhisa Ozaki (Shibaura Institute of Technology) (40/S2-B01 - Salle Bohr) Katsuhisa_Ozaki.mp4 OZAKI_slide_CERN.pdf
16:50	Discussion - Questions (40/S2-B01 - Salle Bohr)
17:10	Closing Session (40/S2-B01 - Salle Bohr)

Choose timezone