Abstract:Multi-objective unconstrained combinatorial optimization problems (MUCO) are in general hard to solve, i.e., the corresponding decision problem is NP-hard and the outcome set is intractable. In this paper we explore special cases of MUCO problems that are actually easy, i.e., solvable in polynomial time. More precisely, we show that MUCO problems with up to two ordinal objective functions plus one real-valued objective function are tractable, and that their complete nondominated set can be computed in polynomial time. For MUCO problems with one ordinal and a second ordinal or real-valued objective function we present an even more efficient algorithm that applies a greedy strategy multiple times.

What problem does this paper attempt to address?

This paper attempts to address the computational complexity of the multi - objective unconstrained combinatorial optimization problem (MUCO). Specifically, the author explores MUCO problems in certain special cases, which can be effectively solved by polynomial - time algorithms. The following are the core problems and solutions in the paper: ### Core Problems in the Paper 1. **Complexity of the Multi - objective Unconstrained Combinatorial Optimization Problem**: - In general, MUCO problems are difficult to solve, that is, their corresponding decision problems are NP - hard, and the result sets are infeasible. - The author aims to identify some special instances of MUCO problems that can be solved in polynomial time. 2. **Specific Types of MUCO Problems**: - It has been found that MUCO problems with one real - valued objective function and at most two ordinal objective functions are tractable, that is, they can be solved in polynomial time. - For MUCO problems with one ordinal objective function and another ordinal or real - valued objective function, the author proposes a more efficient algorithm based on the greedy strategy. ### Solutions 1. **Polynomial - Time Algorithm**: - The author proves that for MUCO problems with one real - valued objective function and at most two ordinal objective functions, the complete non - dominated set can be calculated in polynomial time. - The specific method is to solve a series of ε - constraint scalarization problems, which can be transformed into integer linear programming (ILP) problems, and the constraint matrix is totally unimodular, so it can be solved in polynomial time. 2. **Efficient Algorithm Based on the Greedy Strategy**: - For MUCO problems with one ordinal objective function and another ordinal or real - valued objective function, the author proposes an algorithm based on the greedy strategy. - These algorithms utilize the properties related to the partition matroid, so that the optimal solution can be found in polynomial time. ### Key Conclusions - **Polynomial - Time Complexity**: The author proves that in some special cases, MUCO problems can be solved in polynomial time, which provides theoretical support for solving multi - objective optimization problems in practical applications. - **Efficient Algorithm**: In particular, for MUCO problems with one ordinal objective function and another ordinal or real - valued objective function, the algorithm based on the greedy strategy provides a more efficient solution. ### Example To better understand these concepts, the author provides a small - scale example, showing how to classify elements into different ordinal categories and represent the objective function by an incremental count vector. This example illustrates that in a minimization problem, one solution may be superior to another, while in a maximization problem, the opposite is true. ### Summary By introducing specific types of ordinal objective functions, this paper successfully reduces the computational complexity of the multi - objective unconstrained combinatorial optimization problem, and proposes polynomial - time algorithms and efficient algorithms based on the greedy strategy to solve these special problems. This provides an important theoretical basis and practical tool for solving multi - objective optimization problems in practical applications.

On the Computational Complexity of Multi-Objective Ordinal Unconstrained Combinatorial Optimization

Polyhedral Feasible Set Computation of MPC-based Optimal Control Problems.

Second-order Karush/Kuhn–Tucker conditions and duality for constrained multiobjective optimization problems

Constrained Ordinal Optimization—A Feasibility Model Based Approach

Solving efficiently the 0–1 multi-objective knapsack problem

Probing the Pareto front of a large-scale multiobjective problem with a MIP solver

Output-sensitive Complexity of Multi-Objective Integer Network Flow Problems

Optimization on Pareto sets: On a theory of multi-objective optimization

The Multi-Objective Polynomial Optimization

A penalty decomposition approach for multi-objective cardinality-constrained optimization problems

Cardinality-Constrained Multi-Objective Optimization: Novel Optimality Conditions and Algorithms

Handling Constrained Multiobjective Optimization Problems With Constraints in Both the Decision and Objective Spaces

Multiobjective Optimization of Non-Smooth PDE-Constrained Problems

Tackling Prevalent Conditions in Unsupervised Combinatorial Optimization: Cardinality, Minimum, Covering, and More

A simple, efficient and versatile objective space algorithm for multiobjective integer programming

A new geometric approach to multiobjective linear programming problems

Multitype Integer Monoid Optimization and Applications

Algorithmically Fair Maximization of Multiple Submodular Objective Functions

Relaxations and Duality for Multiobjective Integer Programming

Combinatorial Secretary Problems with Ordinal Information