Diversifying agent's behaviors in interactive decision models

Modeling other agents' behaviors plays an important role in decision models for interactions among multiple agents. To optimize its own decisions, a subject agent needs to model what agents act simultaneously uncertain environment. However, modeling insufficiency occurs when the are competitive and cannot get full knowledge about Even collaborative, they may not share their true due privacy concerns. Most of recent research still assumes that have common environments has behavior mind. Consequently, resulting techniques applicable many practical problem domains. In this article, we investigate into diversifying agent's before interactions. The challenges lie generating measuring new Starting with prior behaviors, use linear reduction technique extract representative behavioral features from known behaviors. We subsequently generate by expanding propose two diversity measurements select top- K $K$ demonstrate performance well-studied selection embarks study unknown multiagent making inspires investigation This will contribute intelligent systems dealing unknowns open artificial intelligence world.