Machine Learning for Social Multiparty Human--Robot Interaction

Abstract

We describe a variety of machine learning techniques that are being applied to social multi-user human-robot interaction, using a robot bartender in our scenario. We first present a data-driven approach to social state recognition based on supervised learning. We then describe an approach to social skills execution-i.e., action selection for generating socially appropriate robot behaviour-which is based on reinforcement learning, using a data-driven simulation of multiple users to train execution policies for social skills. Next, we describe how these components for social state recognition and skills execution have been integrated into an end-to-end robot bartender system, and we discuss the results of a user evaluation. Finally, we present an alternative unsupervised learning framework that combines social state recognition and social skills execution, based on hierarchical Dirichlet processes and an infinite POMDP interaction manager. The models make use of data from both human-human interactions collected in a number of German bars and human-robot interactions recorded in the evaluation of an initial version of the system.

A Systematic Review of Adaptivity in Human-Robot Interaction

“…Keizer et al, 2014 [56] programmed an iCAT robot that played the role of a socially aware bartender robot. The robot was capable of detecting customer, tracking multiple customers, and taking their orders.…”

The MuMMER Project: Engaging Human-Robot Interaction in Real-World Public Spaces

“…In this project, we will advance the state-of-the-art in interaction management in two directions. First, we will apply current state-of-the-art statistical models [13,15,21] to the new scenarios and tasks that arise in the context of engaging, entertaining, socially appropriate human-robot dialogue interaction. Second, we will scale up machine learning techniques to support robust human interaction with a robot in noisy, populated spaces.…”

Introduction to the Special Issue on Machine Learning for Multiple Modalities in Interactive Systems and Robots

“…This function is known as a classifier when the labels are discrete and as a regressor when the labels are continuous. All articles in this special issue make use of classifiers to predict events during human-machine interactions [Ngo et al 2014;Benotti et al 2014;Keizer et al 2014;Cuayáhuitl et al 2014]. -In contrast to supervised learning that makes use of direct feedback, reinforcement learning makes use of indirect feedback typically based on numerical rewards given during the interaction, and the goal is to maximize them in the long run.…”

“…fetching and delivering an object or playing a game). This form of learning is applied in this special issue to guide the behaviour of interactive robots as described in [Keizer et al 2014] and [Cuayáhuitl et al 2014]. -While both supervised and reinforcement learning assume some supervision at learning time, either in the form of labels or rewards, unsupervised learning does not require such information.…”

“…The task of an unsupervised learning algorithm is thus to uncover hidden structure in unlabelled data. Examples of this form of learning include clustering and non-parametric Bayesian learning, which are applied to multimodal interaction in this special issue by [Keizer et al 2014]. …”