Project description
A novel way of solving real-world problems
Submodularity is a property of set functions with deep theoretical consequences and far-reaching applications. With many real-world applications, submodular function maximisation affects many aspects of daily life – from pollution detection to gang violence reduction. With so many real-world applications, problems with a submodal objective have been the focus of intense research. The EU-funded SUBMODULAR project will develop a new theoretical algorithmic foundation for submodular maximisation. The aim will be to enable us to completely resolve basic optimisation problems and facilitate the design of fast and easily implementable algorithms that can be used in practice.
Objective
Submodularity is a fundamental mathematical notion that captures the concept of diminishing returns, and is prevalent in many areas of science and technology. From pollution detection and gang violence reduction, to active influence spreading in social networks (as recent controversial events suggest), submodular maximization directly affects multiple aspects of our daily lives. With a constantly growing number of real-world applications, it is no surprise that combinatorial maximization problems with a submodular objective have been the focus of intense theoretical and practical research for more than a decade. Unfortunately, many of the fundamental submodular maximization problems remain unresolved and lack fast and simple to implement algorithms that can be used in practice. A main reason for this state is that the current algorithmic toolkit used to solve submodular maximization problems is insufficient.
This research proposal addresses the theoretical foundations of submodular maximization. Our overarching goal is to enrich the algorithmic toolkit and devise new algorithmic approaches that can be broadly applied to fundamental problems in submodular maximization and combinatorial optimization. A successful outcome of this research project will be a new theoretical algorithmic foundation for submodular maximization, that will enable us to completely resolve basic optimization problems and facilitate the design of fast and simple to implement algorithms that can be used in practice. The combination of the latter with ongoing practical research on the numerous uses of submodularity, could lead to a significant revolution in our ability to identify and solve the many real-world applications of submodular maximization.
Fields of science
Programme(s)
Topic(s)
Funding Scheme
ERC-STG - Starting GrantHost institution
32000 Haifa
Israel