Emerging Scholars Seminars | Bakar Institute of Digital Materials for the Planet

The BIDMaP Emerging Scholars Seminar is a dynamic, biweekly forum that brings together emerging researchers—graduate students and postdoctoral scholars—at the cutting edge of physics, chemistry, and materials science to explore how artificial intelligence is reshaping these fields. Each session features a scholar presenting their latest research, with a focus on how AI is transforming our ability to understand, design, and manipulate complex materials and systems.

This seminar offers a unique platform for cross-disciplinary dialogue, where experts from physics, materials science, and computational research converge to share novel developments and applications in materials simulations and AI-driven approaches. The goal is to foster collaboration between BIDMaP postdoctoral fellows and researchers in related fields, sparking fresh ideas and pushing the boundaries of current scientific inquiry.

Upcoming Seminars

Feb

Gentlest Ascent Dynamics for Rapid Exploration of Energy Landscapes

The exploration of molecular configuration space is often hindered by thermally activated barriers on complex, high-dimensional potential energy surfaces, making conventional molecular dynamics (MD) simulations inefficient for accessing rare yet important transitions. Enhanced sampling techniques based on collective variables (CVs) are commonly...

Past Seminars

Jan

Moving beyond copper: A search for next generation electrocatalyst materials for CO 2 conversion

Electrochemical CO 2 reduction holds promise for an efficient and decentralized route towards energy and climate resilience by enabling the conversion of CO 2 into value-added chemicals and fuels at mild temperatures and pressures. Early systematic studies by Hori et al. in the mid-1980s identified...

Jan

Transition Path Sampling with Generative Models using Action Minimization

Transition path sampling (TPS), which involves finding probable paths connecting two points on an energy landscape, remains a challenge due to the complexity of real-world atomistic systems. Current machine learning approaches use expensive, task-specific, and data-free training procedures, limiting their ability to benefit from high-quality...

Dec

Accelerating electronic structure calculations using deep neural networks

Artificial intelligence (AI) and big data are becoming deeply integrated into modern scientific research, giving rise to an emerging research field of ab initio AI, which applies state-of-the-art AI techniques to help solve long-standing bottlenecks of ab initio calculations—most notably the tradeoff between accuracy and efficiency. In this talk, Dr. Xiaoxun Gong will discuss the recently developed DeepH method [1–4]...

Nov

Rhys Bunting: Bridging theory to experiment modeling physically more representative systems

Rhys Bunting traces the evolution of atomistic modeling from its molecular dynamics origins to today’s AI-enhanced methods and explores how computation can better capture the complexity of real experimental systems. His talk highlights approaches that integrate diverse modeling techniques to describe catalytic and physical phenomena...

Nov

Andrea Giunto: Accelerating Experimental Material Science with AI and Automation: Compositional Characterization of Powders

Andrea Giunto discusses how AI and automation are accelerating experimental materials science by bridging the gap between computational predictions and real-world synthesis. His talk focuses on automated characterization using SEM-EDS to enable large-scale...

Oct

Lixin Lu: Accurate and Efficient Electronic Structure Methods for Accelerating Discovery

Dr. Lixin Lu presents advances in accurate and efficient electronic structure methods for complex chemical systems, with a focus on heavy elements. Her talk highlights developments in relativistic multi reference methods, GPU-accelerated approaches, and DFT corrected strategies that recover electron correlation efficiently. Through case studies, she demonstrates how these...

Sep

Etienne Palos: Ab Initio and Data-Driven Methods for Many-Body Molecular Simulation

Computer simulation of many-body chemical systems has advanced significantly in recent years through data-driven modeling. Generally, we classify the role of data-driven and machine-learning techniques into (i) accelerating simulation, amplifying the accessible length(time) scales, and (ii) improving method accuracy, reducing the discrepancy...

May

May. 6, 2025 - Jérémie Klinger: Designing optimal control strategies in Stochastic Thermodynamics

The connection between measure-theoretic optimal transport and dissipative nonequilibrium dynamics provides a framework for quantifying nonequilibrium control costs, leading to a geometric interpretation of control and a collection of thermodynamic speed limits. However, realizing such optimal control strategies is challenging both in laboratory...

Apr

Asmita Jana - Nanoconfinement in metal/ligand electrocatalysts: how is it created and why does it reduce CO 2 efficiently?

Enhancing the performance of electrocatalysts for applications like carbon conversion is important for making that technology viable. Nanoconfinement is a promising strategy for doing so. Understanding how it is created and why it enhances reaction rates can lead to a better characterization of nanoconfinement– the...

Mar

Trinidad Novoa - Hydrogen-rich superconductors: efficient estimations of Tc from real-space descriptors

The search for high-temperature superconductors has advanced significantly in the last ten years, with many hydrogen-rich compounds exhibiting superconductivity above 200 K. However, their stability at extreme pressures limits practical applications. Theoretical prediction of new compounds is hindered by expensive calculations of critical temperatures (Tc)...

Mar

Andrew Smith – Simple generalizations in stereoselective catalysis enable the algorithmic discovery of new enantioselective solutions

Structure-activity relationships describe trends between chemical features and subsequent function. As the modeled activity becomes more complex, the relationships drawn often rely on more specific features, inhibiting the realization of general chemical intuition. Notably, trends in...

Feb

Hannes Stärk and Bowen Jing – Dirichlet Flow Matching with Applications to DNA Sequence Design

We develop Dirichlet flow matching on the simplex based on mixtures of Dirichlet distributions as probability paths. In this framework, we derive a connection between the mixtures' scores and the flow's vector field that allows for classifier and classifier-free guidance. On DNA sequence generation tasks...

Feb

Avijit Shee – A static quantum embedding scheme based on coupled cluster theory

We develop a static quantum embedding scheme that utilizes diﬀerent levels of approximations to coupled cluster (CC) theory for an active fragment region and its environment. In this approach, we solve the local fragment problem using a high-level CC method and address the environment problem with a lower-level Møller–Plesset (MP) perturbative...

Jan

Jingyang Wang - Elucidating gas reduction effects of organosilicon additives in lithium-ion batteries

Renewable energy technologies such as lithium-ion batteries have become indispensable in our everyday lives. As current battery research pushes for higher performance and energy density, the operational safety of batteries becomes paramount. Under high voltages and temperatures, nonaqueous...

Dec

Yuanran Zhu – Operator-learning approach for predicting nonequilibrium Green's function dynamics

Operator learning has undergone rapid development in recent years and gradually emerged as a transformative machine-learning paradigm. In this talk, I will provide an overview of the operator-learning framework and its applications to address modern computational challenges, with a focus on quantum many-body dynamics. Using...