The BIDS Machine Learning and Science Forum (formerly the Berkeley Statistics and Machine Learning Forum) was launched in Spring 2018 and currently meets biweekly (during the spring and fall semesters) to discuss current applications across a wide variety of research domains in the physical sciences and beyond. These active sessions bring together domain scientists, statisticians, and computer scientists who are either developing state-of-the-art methods or are interested in applying these methods in their research. This Forum is hosted by BIDS Faculty Affiliate Uroš Seljak (professor of Physics at UC Berkeley) and BIDS Research Affiliate Ben Nachman (Physicist at Lawrence Berkeley National Laboratory), and organized by Vanessa Böhm, Aditi Krishnapriyan, Vinicius Mikuni, Mariel Pettee, and Shashank Subramanian. All interested members of the UC Berkeley and Berkeley Lab communities are welcome and encouraged to attend. To receive email notifications about upcoming meetings, or to request more information, please contact berkeleymlforum@gmail.com.
Machine Learning and Science Forum - Fall 2021
Date: Biweekly on Mondays, September 13 through December 6, 2021
Time: 11:00 AM - 12:00 PM Pacific Time
Location: Participate remotely using this Zoom link
- September 13: Point cloud applications to collider physics, Vinicius Mikuni, NERSC
- September 27: ML Methods for Particle Physics & Choreography, Mariel Pettee, LBNL
- October 11: Data-driven and data-assisted modeling for applications in fluid dynamics and geophysics, Jaideep Pathak, NERSC
- October 25: ML for Social Good in Practice: from Robust Optimization to Broader Impacts in Education, Serena Wang, UC Berkeley
- November 8: Modeling and Simulation Tools for Industrial and Societal Research Applications: Digital Twins and Genome-based Machine-learning, Tarek Zohdi, UC Berkeley
- November 22: Data ethics in government services: putting accountable humans in-the-loop, Eric Giannella, Code for America
- December 6: Deep Generative Modeling of Astronomical Time Series Data, Kyle Boone, University of Washington
- December 13: Title TBA, Eli Ben-Michael, Harvard
Sessions Archive
2021
- January 25: AI at the exascale, Debbie Bard, NERSC
- February 8: A Differential Geometry Perspective on Orthogonal Recurrent Models, Omri Azencot, Ben-Gurion Univeristy
- February 22: Noisy Recurrent Neural Networks, Soon Hoe Lim, KTH
- March 8: Self-Supervised Representation Learning for Astronomical Images, George Stein, BCCP and Berkeley Lab
- March 29: Understanding overparameterized neural networks, Jascha Sohl-Dickstein, Google Brain (Rescheduled from March 22)
- April 5: An ML Control System for the Fermilab Booster, Christian Herwig, Fermilab
- April 19: Improving neural network based equations solvers, Akshunna Shaurya Dogra, UC Berkeley
- May 3: HypE: Learning Knowledge Graph Entity Representations in Hyperbolic Space, Nikhil Rao, Senior ML Scientist, Amazon
- May 17: Session Cancelled, Alejandro Queiruga, Google Research
- May 31: No meeting, Memorial Day Holiday
- June 14: Training stochastic differential equation models using ordinary differential equations, Liam Hodgkinson, Statistics, UC Berkeley
2020
- Jan 27: Reinforcement Learning to Control Quantum Systems, Marin Bukov, Physics, UC Berkeley
- Feb 10: Probabilistic programming at scale for science, Wahid Bhimji, NERSC
- Feb 24: Deep learning for PDEs, and scientific computing with JAX, Stephan Hoyer, Google
- March 9: BoltGAN Machines and Applications in Medicine, Charles K. Fisher, Unlearn.AI
- March 23: Deep generative models for scientific applications, Vanessa Böhm, Berkeley Center for Cosmological Physics
- April 6: Estimating Gradients of Distributions for Generative Modeling, Yang Song, Computer Science, Stanford
- April 20: Robust Mean Estimation in Nearly-Linear Time, with Applications to Outlier Removal, Sam Hopkins, EECS, UC Berkeley
- May 4: Representation Learning for Particle Collider Events, Jack Collins, Physics, SLAC
- May 11: Over-fitting in Modern Supervised Learning: Memorization, Interpolation, and Decomposition of Errors, Jason Rocks, Physics, Boston University
- June 1: Identifying topological order through unsupervised machine learning, Joaquin Rodriguez Nieva, Physics, UC Berkeley
- Sept 14: Comprehension is compression: understanding neural networks through pruning and the lottery ticket hypothesis, Michela Paganini, Facebook AI Research (FAIR)
- Sept 28: FlowPM: Distributed TensorFlow Implementation of Cosmological N-body Solver, Chirag Modi, Berkeley Center for Cosmological Physics
- Oct 12: Machine-Learned Epidemiology, Adam Sadilek, Google Research
- Oct 26: The Large Learning Rate Phase of Deep Learning, Yasaman Bahri, Research Scientist, Google Brain
- Nov 9: First-class machine learning model for Science: Graph Neural Network, Xiangyang Ju, Lawrence Berkeley National Laboratory
- Nov 23: Set and Sequence Machine Learning for Particle Identification at the Large Hadron Collider, Nicole Hartman, SLAC
- Dec 7: Mutual Information Estimation for Tensor Network Machine Learning, Ian Convy, UC Berkeley
2019
- Feb 4: The Information Bottleneck principle in Linguistics — Efficient coding and language evolution: the case of color naming, Noga Zaslavskya, PhD candidate, Center for Brain Sciences, Hebrew University
- March 4: JUNIPR: a framework for unsupervised and interpretable machine learning in particle physics, Anders Andreassen, Postdoctoral Fellow, Physics, LBL
- March 18: Power of gradients and accept-reject step in MCMC algorithms, Raaz Dwivedi, Graduate Student, EECS, UC Berkeley
- April 1: Learning on unstructured spherical grids, Max Jiang, PhD Candidate, UC Berkeley; NERSC Data Analytics Group
- April 15: Interpretable machine learning - what does it actually mean?, Jamie Murdoch, PhD student, Statistics, UC Berkeley
- Sept 9: The Recurrent Inference Machine: applications to Astronomical and Medical imaging, François Lanusse, BIDS-FODA Data Science Fellow, BCCP
- Sept 23: Machine learning for approximating sub-grid physics in electromagnetic geophysics, Lindsey Heagy, UC Berkeley Statistics
- Oct 7: Neural Ordinary Differential Equations, Amir Gholami, BIDS-FODA Data Science Fellow, Berkeley AI Research Lab
- Oct 21: Uncertainties with Neural Networks: Methods and Applications, François Lanusse, BIDS-FODA Data Science Fellow, BCCP
- Nov 4: Anomaly Detection Meets Deep Learning, François Lanusse, BIDS-FODA Data Science Fellow, BCCP
- Nov 18: The role of machine learning in building an earthquake disaster platform, Qingkai Kong, Assistant Researcher, Berkeley Seismology Lab
- Dec 2: Probing Dark Matter with Strong Gravitational Lensing, Simon Birrer, Postdoctoral Fellow, Kavli Institute for Particle Astrophysics and Cosmology (KIPAC), Stanford University
- Dec 16: A family of algorithms for interpreting manifold embedding coordinates in molecular dynamics data, Samson Koelle, IGERT Graduate Student, eScience Institute, University of Washington
2018
- Sept 24: Applications of ML for High Energy Physics, Ben Nachman, LBNL Physics
- Oct 1: Deep Learning in Bio-imaging, Henry Pinkard, PhD student, Computational Biology, UC Berkeley
- Oct 8: Fundamentals of graph theory and application to population migrations, Wooseok Ha, Postdoctoral Fellow, FODA institute, UC Berkeley
- Oct 15: Latest development in Recurrent Neural Networks, Ravi Krishna, EECS, UC Berkeley
- Oct 29: Adaptive Gaussian process surrogates for Bayesian inference, Timur Takhtaganov, LBNL
- Nov 26: Deep Learning with symmetries, Tess Smidt
- Dec 10: Sampling vs Optimization, Yian Ma, UC Berkeley Statistics
