Submodular Maximization Through Barrier Functions
This work addresses optimization problems in machine learning and data science, such as recommendation systems and summarization, by providing improved theoretical guarantees and practical performance, though it is incremental as it builds on existing submodular maximization methods.
The paper tackles constrained submodular maximization by introducing a technique inspired by barrier functions, achieving a state-of-the-art approximation factor of (k+1+ε) for monotone submodular functions under combined constraints, with extensive evaluation on real-world applications like movie recommendation and summarization tasks.
In this paper, we introduce a novel technique for constrained submodular maximization, inspired by barrier functions in continuous optimization. This connection not only improves the running time for constrained submodular maximization but also provides the state of the art guarantee. More precisely, for maximizing a monotone submodular function subject to the combination of a $k$-matchoid and $\ell$-knapsack constraint (for $\ell\leq k$), we propose a potential function that can be approximately minimized. Once we minimize the potential function up to an $ε$ error it is guaranteed that we have found a feasible set with a $2(k+1+ε)$-approximation factor which can indeed be further improved to $(k+1+ε)$ by an enumeration technique. We extensively evaluate the performance of our proposed algorithm over several real-world applications, including a movie recommendation system, summarization tasks for YouTube videos, Twitter feeds and Yelp business locations, and a set cover problem.