October 5 – November 23, 2021
Quantum Monte Carlo Virtual Workshop 2021
Every Tuesday beginning October 5, 2021 for 8 weeks
11am Eastern Time / 8am Pacific Time
Two hours of lectures and hands-on examples each week
Registration deadline: Friday, September 10, 2021
Quantum Monte Carlo (QMC) methods are among the most accurate electronic structure methods, providing an important computational tool for performing many-body calculations for a broad range of electronic systems, from weakly bound molecules to strongly correlated solids. QMC methods can be used directly to access many-body properties of materials and molecules with high accuracy, to help validate more approximate electronic structure methods, or to provide reference data for machine-learning and artificial intelligence applications. QMC methods are consequently particularly in demand for van der Waals and strongly correlated materials where there are many competing interactions. Interest is also increasing rapidly due to algorithmic and capability improvements, increased ease of use, and greater computational affordability.
This multi-week virtual workshop will provide an introduction to these methods, hands-on examples of molecular and solid-state systems, and cover some recent developments. The extended format maximizes flexibility and allows for broader participation. It is particularly suited to students, postdocs, staff, and professors looking to learn about and use these methods either on their own or alongside quantum chemical and density functional theory-based approaches. Some general knowledge of electronic structure or quantum chemical methods is assumed. The intended outcome is that by the end of the workshop, attendees will be able to setup and run their own research calculations.
The workshop is organized by the Center for Predictive Simulation of Functional Materials.
- Quantum Monte Carlo methods
- Modern work tools and automation
- Molecular calculations
- Solid-state calculations
- Incorporating spin-orbit in Quantum Monte Carlo calculations
- Recent research applications
- Paul Kent (Oak Ridge National Laboratory)
- Anouar Benali (Argonne National Laboratory)
- Jaron Krogel (Oak Ridge National Laboratory)
- Ye Luo (Argonne National Laboratory)
- Lubos Mitas (North Carolina State University)
- Brenda Rubenstein (Brown University)
- Luke Shulenburger (Sandia National Laboratory)
Registration is currently open and ongoing, with the initial deadline of Friday, September 10. Please register today! There is no cost to register or participate in the workshop. Registrations after the deadline and during the workshop are welcome but immediate access may not be possible.
While the concepts covered will be general and applicable to all Quantum Monte Carlo codes, the workshop is centered around the use of the QMCPACK code, QMCPACK Manual. Extensive use will be made of the modern work on software NEXUS User Guide, to very efficiently coordinate the different steps of performing QMC calculations. During the workshop, for trial wavefunction calculation, Python-based Simulations of Chemistry Framework (PySCF) will mainly be used for molecular systems and QuantumESPRESSO (QE) for solid-state systems. Attendees should have at least basic familiarity with standard electronic structure methods such as density functional theory and/or wavefunction based quantum chemical approaches.
Each week there will be approximately two hours of presentations and hands-on examples. The virtual workshop will be held using Zoom videoconferencing, with sessions recorded and subsequently uploaded to YouTube. Recordings from workshops held in previous years are available.
Virtual machine images will be provided to run the examples on participants computers. Tutorials will also be given on installing the software and workshop files directly, e.g. on a local workstation or computational cluster. Additional support will be given via a dedicated Slack channel and via an additional support videoconference session each week.
Agenda and presentation materials
(Last updated 23 August 2021)
Lectures and hands-on are 11:00 a.m. (EST) each Tuesday.
An on-demand support session will run each Thursdays at 11:00 a.m. (EST).
Week 1: Workshop Introduction. Running the workshop virtual machine, installing QMCPACK.
Week 2: Fundamentals of Quantum Monte Carlo.
Week 3: Computational Workflows with NEXUS. Statistical analysis.
Week 4: Molecular Calculations.
Week 5: Solid State Calculations.
Week 6: Pseudopotentials. Charge densities and additional observables.
Week 7: Guest lecture and research presentations.
Week 8: Spin-orbit calculations. Next steps and workshop wrap-up.
This workshop is supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division, as part of the Computational Materials Sciences Program and Center for Predictive Simulation of Functional Materials at ORNL.