In less than two weeks, several Berkeley Lab machine learning experts head to New Orleans for the 37th Conference on Information Processing Systems (NeurIPS 2023), held this year from December 10-16, 2023. This multi-track interdisciplinary annual meeting includes invited talks, demonstrations, symposia, and oral and poster presentations of refereed papers. A professional exposition will occur alongside the conference, focusing on machine learning in practice, a series of tutorials, and topical workshops that provide a less formal setting for exchanging ideas.

Below is a continually updated day-by-day guide to NeurIPS 2023 programming featuring Berkeley Lab staff.

All times are Central Standard.

INVITED TUTORIALS

MONDAY, DEC. 11, 2023
Time Title Author(s)/Presenter(s)
7:45 – 10:15 a.m.
CST		 Recent and Upcoming Developments in Randomized Numerical Linear Algebra for ML Michael Mahoney (Berkeley Lab), Michal Derezinski (University of Michigan)
7:45 – 10:15 a.m.
CST		 Contributing to an Efficient and Democratized Large Model Era James Demmel (UC Berkeley / Berkeley Lab), Yang You (National University of Singapore)

Papers Presented in Poster Sessions
DecEMBER 12 – 14

POSTER SESSION 1: TUESDAY, DEC. 12, 2023
Time Title Presenter(s)
10:45 a.m. -12:45 p.m.
CST		 Toward Foundation Models for Scientific Machine Learning: Characterizing Scaling and Transfer Behavior Shashank Subramanian, Peter Harrington, Kurt Keutzer, Wahid Bhimji, Dmitriy Morozov, and Michael Mahoney (Berkeley Lab); Amir Gholami (UC Berkeley)

POSTER SESSION 3: WEDNESDAY, DEC. 13, 2023
Time Title Presenter(s)
10:45 a.m. -12:45 p.m.
CST		 Big Little Transformer Decoder Michael Mahoney (Berkeley Lab), Sehoon Kim, Karttikeya Mangalam, Suhong Moon, Jitendra Malik, Amir Gholami, Kurt Keutzer (UC Berkeley)

POSTER SESSION 4: WEDNESDAY, DEC. 13, 2023
Time Title Presenter(s)
5:00 – 7:00 p.m.
CST		 Fast Exact Leverage Score Sampling from Khatri-Rao Products with Applications to Tensor Decomposition Osman Asif Malik, Aydın Buluç (Berkeley Lab), Vivek Bharadwaj (Berkeley Lab / UC Berkeley), Riley Murray (Sandia National Lab/Berkeley Lab), James Demmel (Berkeley Lab/UC Berkeley), Laura Grigori (EPFL/INRIA Paris)
5:00 – 7:00 p.m.
CST		  A Heavy-Tailed Algebra for Probabilistic Programming Michael Mahoney (Berkeley Lab), Feynman Liang (UC Berkeley), Liam Hodgkinson (University of Melbourne)

POSTER SESSION 5: THURSDAY, DEC. 14, 2023
Time Title Presenter(s)
10:45 a.m. -12:45 p.m.
CST		  When are Ensembles Effective? Michael Mahoney (Berkeley Lab), Ryan Theisen, Hyunsuk Kim (UC Berkeley), Yaoqing Yang (Dartmouth College), Liam Hodgkinson (University of Melbourne)
10:45 a.m. -12:45 p.m.
CST		 Temperature Balancing, Layer-wise Weight Analysis, and Neural Network Training Charles Martin, Michael Mahoney (Berkeley Lab), Yefan Zhou (International Computer Science Institute), Tianyu Pang (SAIL), Keqin Liu (Georgetown University), Yaoqing Yang (Dartmouth College)

INVITED WORKSHOPS

FRIDAY, DEC. 15, 2023
Machine Learning and the Physical Sciences
8:15 a.m. CST
Hall B2
Workshop Website
Organizers: Benjamin Nachman (Berkeley Lab), Brian Nord (Fermilab), Atilim Gunes Baydin (University of Oxford), Adji Bousso Dieng (Princeton University), Emine Kucukbenli (NVIDIA), Siddharth Mishra-Sharma (MIT / Harvard / IAIFI), Kyle Cranmer (University of Wisconsin), Gilles Louppe (University of Liège), Savannah Thais (Columbia University)
Accepted Workshop Papers with Berkeley Lab Authors:
  • High-dimensional and Permutation Invariant Anomaly Detection with Diffusion Generative Models
    Vinicius Mikuni, Benjamin Nachman (Berkeley Lab)
  • Preparing Spectral Data for Machine Learning: A Study of Geological Classification from Aerial Surveys
    Alex Sim, Brian Quiter, Yuxin Wu, Keshang Wu (Berkeley Lab), Jun Woo Chung, Weijie Zhao (Rochester Institute of Technology)
  • Loss Functionals for Learning Likelihood Ratios
    Shahzar Rizvi, Mariel Pettee, Benjamin Nachman (Berkeley Lab)
Heavy Tails in ML: Structure, Stability, Dynamics
9:00 a.m. CST
Room R02-R05
Workshop Website
Organizers: Michael Mahoney (Berkeley Lab), Mert Gurbuzbalaban (Rutgers), Stefanie Jegelka (MIT), Umut Simsekli (Inria Paris / ENS)

SATURDAY, DEC. 16, 2023
The Symbiosis of Deep Learning  and Differential Equations – III
8:30 a.m. CST
Room 255-257
Workshop Website
Organizers: Ermal Rrapaj (UC Berkeley / Berkeley Lab), Luca Herranz-Celotti (Université de Sherbrooke), Martin Magill (Borealis AI), Winnie Xu (University of Toronto / Google Brain), Qiyao Wei (University of Cambridge), Archis Joglekar (University of Michigan / Syntensor), Michael Poli (Stanford University), Animashree Anandkumar (Caltech)
Third Workshop on Efficient Natural Language and Speech Processing (ENLSP-III): Towards the Future of Large Language Models and their Emerging Descendants
8:30 a.m. CST
Room 206-207
Workshop Website
Accepted Workshop Paper with a Berkeley Lab Author: 

About Computing Sciences at Berkeley Lab

High performance computing plays a critical role in scientific discovery. Researchers increasingly rely on advances in computer science, mathematics, computational science, data science, and large-scale computing and networking to increase our understanding of ourselves, our planet, and our universe. Berkeley Lab's Computing Sciences Area researches, develops, and deploys new foundations, tools, and technologies to meet these needs and to advance research across a broad range of scientific disciplines.