# Interactively shaping robot behaviour with unlabeled human instructions

## TL;DR

This paper introduces TICS, a modular framework enabling humans to shape robot behavior through unlabeled instructions, combining multiple information sources to improve learning efficiency in robotic tasks.

## Contribution

The paper presents a novel framework that integrates unlabeled human instructions with reward and feedback for robot learning, bridging reinforcement and supervised learning.

## Key findings

- Accelerates robot task learning
- Reduces the number of teaching signals needed
- Effective in both simulation and real robot experiments

## Abstract

In this paper, we propose a framework that enables a human teacher to shape a robot behaviour by interactively providing it with unlabeled instructions. We ground the meaning of instruction signals in the task-learning process, and use them simultaneously for guiding the latter. We implement our framework as a modular architecture, named TICS (Task-Instruction-Contingency-Shaping) that combines different information sources: a predefined reward function, human evaluative feedback and unlabeled instructions. This approach provides a novel perspective for robotic task learning that lies between Reinforcement Learning and Supervised Learning paradigms. We evaluate our framework both in simulation and with a real robot. The experimental results demonstrate the effectiveness of our framework in accelerating the task-learning process and in reducing the number of required teaching signals.