Interactive Teaching for Conversational AI

Abstract

Current conversational AI systems aim to understand a set of pre-designed requests and execute related actions, which limits them to evolve naturally and adapt based on human interactions. Motivated by how children learn their first language interacting with adults, this paper describes a new Teachable AI system that is capable of learning new language nuggets called concepts, directly from end users using live interactive teaching sessions. The proposed setup uses three models to: a) Identify gaps in understanding automatically during live conversational interactions, b) Learn the respective interpretations of such unknown concepts from live interactions with users, and c) Manage a classroom sub-dialogue specifically tailored for interactive teaching sessions. We propose state-of-the-art transformer based neural architectures of models, fine-tuned on top of pre-trained models, and show accuracy improvements on the respective components. We demonstrate that this method is very promising in leading way to build more adaptive and personalized language understanding models.

The five Is: Key principles for interpretable and safe conversational AI

“…Investing research efforts into interactive learning methods and inquisitiveness routines is likely to be of crucial importance for developing systems that adhere to the principles introduced here. Such methods need not themselves be interpretable (see, e.g., [24]), as long as their final output, i.e. the conversational agent, is interpretable.…”

“…Here, as the fourth key design principle, we argue that interactive learning whereby a person (acting as a tutor) teaches a machine new capabilities [24], [25], should be a primary approach when building conversational agents. Coupled with the inquisitiveness (described below), this method has the advantage of resembling the manner in which humans, not least children, learn new concepts.…”