Talk

• Time/Date: 3pm-4:30pm, Feb 16, 2024
• Place: Room 201
• Title: Affective Cognition Methods and Recent Advances
• Speaker: Prof. Xuefeng Liang, Xidian University & Kyoto University
• Abstract: For a long time, emotion is one of the important criteria to distinguish between human and machine. Currently, AI has shown rapid growth and comprehensive expansion. The "perceptual intelligence" is gradually changing to "cognitive intelligence" with the ability to understand and express, and it has become an inevitable trend to give emotions to machines. With the release of ChatGPT, AI technology has taken a big step forward to Artificial General Intelligence (AGI), but there is still a lot of gap in the perception and generation of emotions. This talk firstly introduce the background of emotion perception, the manifestation of emotion signals, and the traditional methods of emotion analysis. After that, it introduces deep learning-based multimodal sentiment analysis methods, and our latest research results from sentiment ambiguity, sentiment data distribution characteristics, and multimodal network learning methods.
• Flyer