Skip to main content
Go to the home page of the European Commission (opens in new window)
English en
CORDIS - EU research results
CORDIS

Parameterized Algorithms and Polyhedra

Project description

An integrated approach to tackle complex optimisation problems

Polyhedral techniques use geometric shapes – polyhedral – to solve optimisation problems by analysing constraints and finding the best solutions. While these methods are widely used in linear programming and combinatorial optimisation (problems that involve finding the most efficient way to arrange or select items), they are rarely applied in parameterised algorithms, which focus on solving problems efficiently by isolating key parameters. The ERC-funded PARAMLP project will combine the latter with polyhedral techniques to address open questions in combinatorial optimisation. The study will explore new ways to use polyhedral techniques in parameterised algorithms, address gaps in understanding algebraic methods for matroid problems and improve approximation techniques for tasks such as scheduling and fair resource allocation.

Objective

Polyhedral techniques allow the powerhouse of linear programming to be used for the discrete problems in combinatorial optimization. Their impact on commercial solvers, the theory of classical algorithms, as well as approximation algorithms, is well established. Nevertheless, these techniques play only a marginal role in mainstream research on parameterized algorithms. Recent advances, many of which are due to the PI, are showing potential for a revolutionary synergy between parameterized algorithms and polyhedral techniques. This project aims to unlock the full power of this synergy and use it to address several pressing open questions in combinatorial optimization. The work plan consists of three complementary directions:

I. Exploring new frontiers for polyhedral techniques in parameterized algorithms, in particular, the novel usage of proximity bounds to obtain FPT algorithms. We will specifically investigate the open question of whether bin packing in the number of item sizes is in FPT.

II. Closing notorious loose ends in existing research lines, in particular, the lack of polyhedral understanding of algebraic methods in matroid problems.

III. Embracing parameterization to strengthen polyhedral techniques in approximation algorithms. Specifically, we will investigate an innovative technique crucial for understanding the approximability of scheduling on unrelated parallel machines, max-min fair allocation, and directed Steiner tree.

The topics of the project are relevant for a broad spectrum of researchers from both parameterized algorithms and approximation algorithms and have the potential to bring these fields much closer together.

Fields of science (EuroSciVoc)

Programme(s)

Topic(s)

Funding Scheme

HORIZON-ERC - HORIZON ERC Grants

See all projects funded under this funding scheme

Call for proposal

(opens in new window) ERC-2025-STG

See all projects funded under this call

Host institution

SYDDANSK UNIVERSITET
Net EU contribution
€ 1 498 648,00
Address
CAMPUSVEJ 55
5230 Odense M
Denmark
Region
Danmark Syddanmark Fyn
Activity type
Higher or Secondary Education Establishments
Links
Total cost
€ 1 498 648,00

Beneficiaries (1)