BIDMaP is happy to be jointly hosting this year's seminar series with colleagues in computational physical sciences, including collaborators from the departments of Physics and Astronomy, and the Lawrence Berkeley National Laboratory.
Biography: Professor Garnett's primary research interest is Bayesian active learning, with a focus on applications in the natural sciences and engineering. A major theme in his research is automating scientific discovery, broadly interpreted to include both theory and practice and both policy design and modeling...
Biography: Zachary W. Ulissi joined Meta’s Fundamental AI Research lab in 2023 to work on AI for chemistry and climate applications and is based in the San Francisco Bay Area. He is particularly excited about how AI and machine learning methods can enhance various quantum...
Biography: Dr. Ching-Yao Lai and her group attack fundamental questions in ice-dynamics, geophysics, and fluid dynamics by integrating mathematical and machine-learned models with observational data. They use their findings to address challenges facing the world, such as advancing our scientific knowledge of ice dynamics under...
The future of chemistry is self-driving In this talk, I will overview the growing field of self-driving laboratories (SDLs). SDLs are systems that help accelerate the process of scientific discovery or scale-up by employing artificial intelligence and automation for experiment planning and execution. Several SDLs...
Statistical inference is a crucial step in gleaning insights from experimental / observational data to build better physics models to describe our universe. Whether it is to unravel the mysteries of what is happening in the interior of neutron stars, or to uncover the secrets...
The need for new materials to tackle societal challenges in energy and sustainability is widely acknowledged. As demands for performance increase while resource constraints narrow available options, the vastness of composition, structure and process parameter space make the apparently simple questions of where to look...
Composite materials are known for their customizable properties and superior performance characteristics. However, the design of these materials is inherently complex, as it involves navigating through an extensive array of possible material combinations and configurations. In this talk, I will first present novel computational approaches...
Studying low-likelihood high-impact climate events in a warming world requires massive ensembles of hindcasts to capture their statistics. It is currently not feasible to generate these ensembles using traditional weather or climate models, especially at sufficiently high spatial resolution. We describe how to bring the...
Metal-organic frameworks are ultraporous materials that can exhibit selective and cooperative CO2 adsorption chemistry, with potential for future reversible carbon capture applications. While CO2 adsorption enthalpies are relatively well documented, many temperature-dependent and chemical dynamical phenomena related to CO2 and other competing molecular species remain...
The launch of ChatGPT in November 2022 ignited an ongoing worldwide conversation about the possible impacts of Large Language Models (LLMs) on the way we work. There is little doubt that LLMs will significantly influence many people's jobs: one prominent study estimated that about 20%...
The annual BASC Symposium will take place on March 7 and 8, 2024. This year, the theme is “Going with the flow: AI/ML in Atmospheric Science.” The speakers and schedule appear below. The “early-ish bird” registration deadline is this Friday, March 1 st , but...
As scientific data sets become progressively larger algorithms to process this data quickly become more complex. In response Artificial Intelligence (AI) has emerged as a solution to efficiently analyze these massive data sets. Emerging processor technologies such as graphics processing units (GPUs) and field-programmable gate...
Machine learning methods for improving molecular simulations Molecular simulations aim to model the spatiotemporal behavior of atomistic systems throughout biology, chemistry, and materials science. Given the computational burden of running such simulations for long timescales, machine learning force fields, and particularly neural network interatomic potentials...
Current and future weak lensing surveys contain significant information about our universe, but their optimal cosmological analysis is challenging, with traditional analyses often resulting in information loss due to reliance on summary statistics like two-point correlation functions. While deep learning methods offer promise in capturing...
The analysis of particle collisions at the Large Hadron Collider at CERN helps us to understand the fundamental building blocks of our universe. After years of data taking, the extraction of fundamental insights often requires intricate data analyses. Due to the large volume and complex...
Some of the most powerful techniques developed in ML are rooted in physics, such as MCMC, Belief Propagation, and Diffusion based Generative AI. We have recently witnessed that the flow of information has also reversed, with new tools developed in the ML community impacting physics...
Atomic systems (molecules, crystals, proteins, etc.) are naturally represented by a set of coordinates in 3D space labeled by atom type. This poses a challenge for machine learning due to the sensitivity of coordinates to 3D rotations, translations, and inversions (the symmetries of 3D Euclidean...
Density functional theory (DFT) lies at the heart of all practical applications of theoretical chemistry. However, it is well-known that DFT often provides unsatisfactory descriptions of many important systems: non-equilibrium geometries, such as transition states, and strongly correlated systems, such as transition metals and excited...