Affective Manifolds: Modeling Machine's Mind to Like, Dislike, Enjoy, Suffer, Worry, Fear, and Feel Like A Human

TL;DR

This paper introduces affective manifolds as a way to model emotional states in machines, enabling them to react emotionally and interact more naturally with humans and other machines.

Contribution

It proposes a novel framework combining manifold learning and deep metric learning to model and train emotional states in machines, expanding the concept of machine cognition.

Findings

Affective manifolds can model various emotional states in machines.

Deep metric learning effectively maps input signals to affective states.

Multiple affective manifolds enhance the realism and effectiveness of machine responses.

Abstract

After the development of different machine learning and manifold learning algorithms, it may be a good time to put them together to make a powerful mind for machine. In this work, we propose affective manifolds as components of a machine's mind. Every affective manifold models a characteristic group of mind and contains multiple states. We define the machine's mind as a set of affective manifolds. We use a learning model for mapping the input signals to the embedding space of affective manifold. Using this mapping, a machine or a robot takes an input signal and can react emotionally to it. We use deep metric learning, with Siamese network, and propose a loss function for affective manifold learning. We define margins between states based on the psychological and philosophical studies. Using triplets of instances, we train the network to minimize the variance of every state and have the desired distances between states. We show that affective manifolds can have various applications for machine-machine and human-machine interactions. Some simulations are also provided for verification of the proposed method. It is possible to have as many affective manifolds as required in machine's mind. More affective manifolds in the machine's mind can make it more realistic and effective. This paper opens the door; we invite the researchers from various fields of science to propose more affective manifolds to be inserted in machine's mind.