Reinforcement learning (RL) has become widely adopted in robot control. Despite many successes, one major persisting problem can be very low data efficiency. One solution is interactive feedback, which has been shown to speed up RL considerably. As a result, there is an abundance of different strategies, which are, however, primarily tested on discrete grid-world and small scale optimal control scenarios. In the literature, there is no consensus about which feedback frequency is optimal or at which time the feedback is most beneficial. To resolve these discrepancies we isolate and quantify the effect of feedback frequency in robotic tasks with continuous state and action spaces. The experiments encompass inverse kinematics learning for robotic manipulator arms of different complexity. We show that seemingly contradictory reported phenomena occur at different complexity levels. Furthermore, our results suggest that no single ideal feedback frequency exists. Rather that feedback frequency should be changed as the agent’s proficiency in the task increases.

The number of worldwide inhabitants suffering from visual or hearing impairments reaches billions according to the World Health Organization, making the need for universal access and inclusion in Intelligent Environments (IE) essential. An adaptive Rock-Paper-Scissors application using a simulation of the social robot Haru is presented. The accessibility of the application which covers three modes - where the user able to see and hear, only to see, or only to hear – was verified through a user-study. A multivariate analysis of variance with repeated measures determined that the ratings from the 12 participants differed significantly across the three modes with F(6,6) = 6.823, η2p = .872, p = .017. Results show that users tend to expect applications to be harder to use when suffering from a disability, especially a visual impairment. All modes in the application were deemed acceptable in terms of usability, proving that the multimodality that comes with IE can help in promoting universal access and reducing social exclusion.

No abstract