Learning Adverbs with Spectral Mixture Kernels
Keywords: adverbs, language learning
Abstract: For humans and robots to collaborate more in the real world, robots need to understand human intentions from the different manner of their behaviors. In this study, we focus on the meaning of adverbs which describe human motions. We propose a topic model, Hierarchical Dirichlet Process-Spectral Mixture Latent Dirichlet Allocation, which concurrently learns the relationship between human motions and adverbs by capturing the frequency kernels that represent motion characteristics and the shared topics of adverbs to depict such motions. We trained the model on datasets we made from movies about “walking” and “dancing”, and found that our model outperforms representative neural network models in terms of perplexity score. We also demonstrate our model’s ability to estimate suitable adverbs for a given motion automatically extracted from a movie.
Submission Type: Long Paper (8 Pages)
Archival Option: This is a non-archival submission
Submission Number: 10
Loading