Planning with Submodular Objective Functions
arXiv:2010.11863
Abstract
We study planning with submodular objective functions, where instead of maximizing the cumulative reward, the goal is to maximize the objective value induced by a submodular function. Our framework subsumes standard planning and submodular maximization with cardinality constraints as special cases, and thus many practical applications can be naturally formulated within our framework. Based on the notion of multilinear extension, we propose a novel and theoretically principled algorithmic framework for planning with submodular objective functions, which recovers classical algorithms when applied to the two special cases mentioned above. Empirically, our approach significantly outperforms baseline algorithms on synthetic environments and navigation tasks.
References in corpus (6)
- Online Planning Algorithms for POMDPs
- Anytime Point-Based Approximations for Large POMDPs
- Matterport3D: Learning from RGB-D Data in Indoor Environments
- Variance Reduction Methods for Sublinear Reinforcement Learning
- Towards Nearly-linear Time Algorithms for Submodular Maximization with a Matroid Constraint
- Heterogeneous Measurement Selection for Vehicle Tracking using Submodular Optimization