Affectively Framework: Towards Human-like Affect-Based Agents
This addresses the problem of developing more human-like affect-based agents in game environments for AI and robotics researchers, though it appears incremental as it builds on existing reinforcement learning and affect modeling concepts.
The paper tackles the lack of reinforcement learning frameworks that incorporate human affect models by introducing the Affectively Framework, a set of Open-AI Gym environments that integrate affect into the observation space, with baseline experiments validating its effectiveness.
Game environments offer a unique opportunity for training virtual agents due to their interactive nature, which provides diverse play traces and affect labels. Despite their potential, no reinforcement learning framework incorporates human affect models as part of their observation space or reward mechanism. To address this, we present the \emph{Affectively Framework}, a set of Open-AI Gym environments that integrate affect as part of the observation space. This paper introduces the framework and its three game environments and provides baseline experiments to validate its effectiveness and potential.