Key points are not available for this paper at this time.
With advances in machine learning and artificial intelligence, a considerable impact is brought to all aspects of people’s lifestyles in terms of work, social, and economy. Especially, representation learning, which is one of the most crucial roles of deep learning, is developing rapidly and has been applied to many areas. Representation learning is expected to discover useful features or representations from complex, redundant, and highly variable data, such as images, video, and sensory data. In particular, through representation learning, a machine is available to learn the features rather than use the features. However, technologists have largely ignored emotion and created an often frustrating experience for people, in part because affect has been misunderstood and hard to measure. Emotion is fundamental to human experience, influencing cognition, perception, and everyday tasks such as learning, communication, and even rational decision-making. Although the advanced techniques have considerably provided a lot of intelligent services, it is not adequate to provide affective services, including various unique aspects, e.g., sentiment analysis, emotion recognition, affective interaction, affective computing, and so on. Under the new service paradigm, novel affective services and innovative applications need to be extensively investigated to gain the high potentials brought by representation learning.
Zhang⋆ et al. (Mon,) studied this question.