|Scalable and interpretable data representation for high-dimensional, complex data
|Year of Conference
|Kim, B., K. Patel, A. Rostamizadeh, and J. Shah
|AAAI Conference on Artificial Intelligence (AAAI)
The majority of machine learning research has been focused on building models and inference techniques with sound mathematical properties and cutting edge performance. Little attention has been devoted to the development of data representation that can be used to improve a user's ability to interpret, evaluate and debug machine learning models to solve real-world problems. In this paper, we quantitatively and qualitatively verify that an efficient, accurate and scalable feature-compression method using latent Dirichlet allocation effectively communicates the characteristics of high-dimensional, complex data points. We show that the improvement of a user's interpretability through the use of a topic modeling-based compression technique is statistically significant, according to a number of metrics, when compared with other representations. Also, we find that this representation is scalable --- it maintains alignment with human classification accuracy as an increasing number of data points are shown. In addition, the learned topic layer can semantically deliver meaningful information to users that could potentially aid human reasoning about data characteristics in connection with compressed topic space.