Attending to Emotional Narratives
arXiv:1907.04197
Abstract
Attention mechanisms in deep neural networks have achieved excellent performance on sequence-prediction tasks. Here, we show that these recently-proposed attention-based mechanisms---in particular, the Transformer with its parallelizable self-attention layers, and the Memory Fusion Network with attention across modalities and time---also generalize well to multimodal time-series emotion recognition. Using a recently-introduced dataset of emotional autobiographical narratives, we adapt and apply these two attention mechanisms to predict emotional valence over time. Our models perform extremely well, in some cases reaching a performance comparable with human raters. We end with a discussion of the implications of attention mechanisms to affective computing.
arXiv commentAccepted at IEEE Affective Computing and Intelligent Interaction (ACII) 2019; 6 pages + 1 page ref; 4 figures
References in corpus (5)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Non-Autoregressive Neural Machine Translation
- QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
- Memory Fusion Network for Multi-view Sequential Learning
- Modeling emotion in complex stories: the Stanford Emotional Narratives Dataset