Abstract

Traditional active learning allows a (machine) learner to query the (human) teacher for labels on examples it finds confusing. The teacher then provides a label for only that instance. This is quite restrictive. In this paper, we propose a learning paradigm in which the learner communicates its belief (i.e. predicted label) about the actively chosen example to the teacher. The teacher then confirms or rejects the predicted label. More importantly, if rejected, the teacher communicates an explanation for why the learner ’s belief was wrong. This ex- planation allows the learner to propagate the feedback provided by the teacher to many unlabeled images. This allows a classi fier to better learn from its mistakes, leading to accelerated discriminative learning of visual concepts even with few la- beled images. In order for such communication to be feasible, it is crucial to have a language that both the human supervisor and the machine learner understand. Attributes provide precisely this channel. They are human-interpretable mid-level visual concepts shareable across categories e.g. “furry”, “spacious”, etc. We ad- vocate the use of attributes for a supervisor to provide feedback to a classi fier and directly communicate his knowledge of the world. We employ a straightforward approach to incorporate this feedback in the classi fier, and demonstrate its power on a variety of visual recognition scenarios such as image classi fication and an- notation. This application of attributes for providing classi fiers feedback is very powerful, and has not been explored in the community. It introduces a new mode of supervision, and opens up several avenues for future research.