An end-to-end framework for mixed-integer optimization with data-driven learned constraints.
-
Updated
Sep 4, 2023 - Jupyter Notebook
An end-to-end framework for mixed-integer optimization with data-driven learned constraints.
official repo for the NeurIPS 2022 paper "A Solver-Free Framework for Scalable Learning in Neural ILP Architectures"
Official Code for ICLR 2023 paper "Learning Soft Constraints from Constrained Expert Demonstrations"
Training reliable AI models with neuro-symbolic verification and constraint-guided learning.
NCET (Neural-network Constraint Embedding Toolkit) converts supported PyTorch neural networks into exact mixed-integer linear constraints compiled by CVXPY.
Blog post on safety constraint learning
To associate your repository with the constraint-learning topic, visit your repo's landing page and select "manage topics."