Minimal-envy Conference Paper Assignment: Formulation and a Fast Iterative Algorithm

In conference review management, it is very common that many papers being assigned are not liked by some reviewers due to their specialties or personal interests. In this paper, we consider a reviewer might only envy other reviewers who get a better set of papers than his. In this context, we proposes a new bi-objective optimization assignment model to efficiently and fairly assign papers to many competing reviewers, in which an index that measures the amount of envy is introduced. A fast iterative algorithm is proposed to solve the problem. The experimental results on real datasets shows that the algorithm achieves efficient and fair results and good computing performance, especially on large scale problems. This work is also meaningful for a number of similar allocation or matching problems.