A Deep Generative Model for Interactive Data Annotation through Direct Manipulation in Latent Space
This work addresses the data annotation bottleneck for ML practitioners, but it appears incremental as it builds on existing tools with new analogies and model variants.
The paper tackles the problem of limited annotated data for machine learning by developing a deep generative model for interactive data annotation through direct manipulation in latent space, proposing new analogies and a network model to learn compact graphical representations, with results including identification of candidate model variants for future user studies.
The impact of machine learning (ML) in many fields of application is constrained by lack of annotated data. Among existing tools for ML-assisted data annotation, one little explored tool type relies on an analogy between the coordinates of a graphical user interface and the latent space of a neural network for interaction through direct manipulation. In the present work, we 1) expand the paradigm by proposing two new analogies: time and force as reflecting iterations and gradients of network training; 2) propose a network model for learning a compact graphical representation of the data that takes into account both its internal structure and user provided annotations; and 3) investigate the impact of model hyperparameters on the learned graphical representations of the data, identifying candidate model variants for a future user study.