QUEST

Official Resources

• Homepage: http://physics.wm.edu/~shiwei/quest/ (Research Group)
• Documentation: Available within the package/group resources
• Source Repository: Contact Shiwei Zhang group (William & Mary / Flatiron CCQ)
• License: Academic/Research (typically on request)

Overview

QUEST (Quantum Electron Simulation Toolbox) is a Determinant Quantum Monte Carlo (DQMC) code developed by Shiwei Zhang's group (College of William & Mary, now also Flatiron CCQ). It is designed for high-precision simulations of strongly correlated electron systems, particularly the 2D Hubbard model and its variants. QUEST implements efficient algorithms for ground-state and finite-temperature auxiliary-field QMC (AFQMC), emphasizing numerical stability and control over the sign problem through constrained path methods.

Scientific domain: Determinant QMC, AFQMC, Hubbard models, strongly correlated electrons Target user community: Condensed matter theorists, research groups studying Hubbard physics

Theoretical Methods

• Determinant Quantum Monte Carlo (DQMC)
• Auxiliary-Field Quantum Monte Carlo (AFQMC)
• Constrained Path Monte Carlo (CPMC)
• Ground-state projection
• Finite-temperature grand canonical ensemble
• Hubbard-Stratonovich transformation
• Numerical stabilization techniques

Capabilities (CRITICAL)

• Hubbard Model Solver: Specialized for 2D Hubbard models (single and multi-orbital)
• Sign Problem Control: Implements constrained path approximation to mitigate sign problem
• Numerical Stability: Advanced matrix decomposition and stabilization
• Observables:
  • Energy and double occupancy
  • Magnetic correlations (spin-spin)
  • Superconducting pairing correlations
  • Density matrices
• Lattices: Square, rectangular, and custom geometries
• Performance: Optimized for large-scale production runs

Sources: Shiwei Zhang group publications, "The Quantum Electron Simulation Toolbox (QUEST)"

Inputs & Outputs

• Input formats:

  • Fortran/C-style input parameters (lattice size, interaction U, temperature, etc.)
  • Simulation control parameters (trotter step, number of walkers)

• Output data types:

  • Green's functions
  • Correlation functions
  • Statistical averages and error estimates
  • Checkpoint files

Interfaces & Ecosystem

• Research Code: Often used as a standalone engine or integrated into group-specific workflows.
• Related Tools: Precursor/related to CPMC-Lab (Matlab package) and newer CCQ AFQMC libraries.
• Development: actively maintained within the research group context.

Limitations & Known Constraints

• Availability: Not a public open-source repository like GitHub; typically obtained via collaboration or request.
• Documentation: Research-grade; assumes familiarity with DQMC/AFQMC.
• Scope: Primarily focused on Hubbard-type lattice models.

Performance Characteristics

• Scalability: Designed for large-scale production runs on HPC clusters.
• Stability: Emphasizes numerical control over the sign problem, allowing access to lower temperatures than standard implementations.
• Efficiency: Optimized for 2D lattices.

Comparison with Other Codes

Verification & Sources

Primary sources:

1. QUEST Project Page: http://physics.wm.edu/~shiwei/quest/
2. Shiwei Zhang Publications (e.g., Phys. Rev. Lett., Phys. Rev. B)
3. "Constraint Path Monte Carlo" methodology papers

Confidence: VERIFIED - Leading research code in the field

Verification status: ✅ VERIFIED

• Category: Academic Research Code
• Status: Active (used in diverse publications)
• Developer: Shiwei Zhang Group
• Specialized strength: Controlled accuracy for Hubbard models via Constrained Path MC.