Software & Codes

Our main research outputs, including a range of domain-specific and general-purpose optimization solvers, are available as open-source software. For detailed introductions and user manuals, please refer to the corresponding source code repositories and project homepages.


GPU-based Algorithms

Algorithm Description
PDBO A primal-dual algorithm for solving unconstrained binary optimization problems
MAPE A fast GPU-based method for nonlinear integer programming

AI-Assisted Algorithms

Algorithm Description
IPM-LSTM A learning-based interior point method for nonlinear programming
SymILO A symmetry-aware learning framework for integer linear optimization
ROS A GNN-based relax-optimize-and-sample framework for Max-k-Cut
Predict-and-Search A GNN-guided predict-and-search framework for mixed-integer linear programming

Power Systems Optimization

Algorithm Description
Approximate-and-Optimize An approximate-and-optimize method for security-constrained optimal power flow
Relax-and-Cut A relax-and-cut approach for addressing SCUC
FOM-SCUC A successive fixing approach for SCUC using first-order methods