Human-Guided Object Mapping for Task Transfer

Abstract

When transferring a learned task to an environment containing new objects, a core problem is identifying the mapping between objects in the old and new environments. This object mapping is dependent on the task being performed and the roles objects play in that task. Prior work assumes (i) the robot has access to multiple new demonstrations of the task or (ii) the primary features for object mapping have been specified. We introduce an approach that is not constrained by either assumption but rather uses structured interaction with a human teacher to infer an object mapping for task transfer. We describe three experiments: an extensive evaluation of assisted object mapping in simulation, an interactive evaluation incorporating demonstration and assistance data from a user study involving 10 participants, and an offline evaluation of the robot's confidence during object mapping. Our results indicate that human-guided object mapping provided a balance between mapping performance and autonomy, resulting in (i) up to 2.25× as many correct object mappings as mapping without human interaction, and (ii) more efficient transfer than requiring the human teacher to re-demonstrate the task in the new environment, correctly inferring the object mapping across 93.3% of the tasks and requiring at most one interactive assist in the typical case.

Theory-Based Causal Transfer:Integrating Instance-Level Induction and Abstract-Level Structure Learning

“…More complex theories could potentially account for delayed effects; e.g., when an agent could not find a causal attribute for a particular event, the agent could examine attributes jointly to best explain the causal effect observed. Prior work has examined structural analogies (Hinrichs and Forbus 2011;Zhang et al 2019a;2019b) and object mappings (Fitzgerald, Goel, and Thomaz 2018) to facilitate transfer; these may also be useful to acquire transferable causal knowledge.…”

A Cognitive Reformation

“…illustrates, recent work in human–robot interaction is beginning to address this problem. Fitzgerald et al () describe how a robot not only can build an analogical mapping between the new and a known situation, actively asking questions of the human teacher as needed, but also transform the action model in the familiar situation to the new situation. I think that this kind of research can help inform cognitive science, and, in turn, benefit from input and feedback from cognitive scientists.…”

“…In order to make the robots work, Fitzgerald et al () and Parashar et al () developed representations and processes that offer hypotheses for cognitive science to explore, and cognitive science likely may offer useful inputs and critiques to them. I expect that there are many examples—dozens, scores, perhaps hundreds of examples—not only from AI but also from anthropology and philosophy that could benefit cognitive science as well as benefit from it.…”

Educational Robotics and Robot Creativity: An Interdisciplinary Dialogue

“…Several approaches have been introduced to reduce reinforcement learning time; these include learning through other agent’s advice in a shared environment ( Saunders, 2012 ; Silva and Costa, 2019 ), and learning from human demonstrations ( Argall et al, 2009 ; Fitzgerald et al, 2018 ). Another way to overcome the drawback of time-consuming exploration is to enable machine learning algorithms with the ability to transfer and reuse previously acquired knowledge across tasks using a case-based reasoning approach (CBR) ( Riesbeck and Schank, 1989 ; Kolodner, 2014 ).…”