Stochastic Constraint Programming as Reinforcement Learning
This work addresses the scalability issue in SCP for researchers and practitioners dealing with uncertain constraint-based problems, though it appears incremental as it integrates existing methods.
The paper tackled the challenge of scaling Stochastic Constraint Programming (SCP) to large problems by combining it with Reinforcement Learning (RL), resulting in a hybrid prototype that demonstrates usefulness on SCP problems.
Stochastic Constraint Programming (SCP) is an extension of Constraint Programming (CP) used for modelling and solving problems involving constraints and uncertainty. SCP inherits excellent modelling abilities and filtering algorithms from CP, but so far it has not been applied to large problems. Reinforcement Learning (RL) extends Dynamic Programming to large stochastic problems, but is problem-specific and has no generic solvers. We propose a hybrid combining the scalability of RL with the modelling and constraint filtering methods of CP. We implement a prototype in a CP system and demonstrate its usefulness on SCP problems.