An attention model for the formation of collectives in real-world domains

TL;DR

This paper introduces a novel attention-based model combined with ILP to form collectives of agents in real-world applications, achieving high-quality solutions in domains like ridesharing and cooperative learning.

Contribution

It presents a new approach that transforms collective formation problems into weighted set packing problems using an attention encoder-decoder, solved efficiently by ILP.

Findings

Achieves solutions comparable to domain-specific state-of-the-art methods.

Outperforms recent general approaches based on Monte Carlo tree search.

Effective in real-world applications like ridesharing and team formation.

Abstract

We consider the problem of forming collectives of agents for real-world applications aligned with Sustainable Development Goals (e.g., shared mobility, cooperative learning). We propose a general approach for the formation of collectives based on a novel combination of an attention model and an integer linear program (ILP). In more detail, we propose an attention encoder-decoder model that transforms a collective formation instance to a weighted set packing problem, which is then solved by an ILP. Results on two real-world domains (i.e., ridesharing and team formation for cooperative learning) show that our approach provides solutions that are comparable (in terms of quality) to the ones produced by state-of-the-art approaches specific to each domain. Moreover, our solution outperforms the most recent general approach for forming collectives based on Monte Carlo tree search.