Natural methods for robot task learning: instructive demonstrations, generalization and practice

By Monica N. Nicolescu, Maja J. Mataric. Adaptive Agents and Multi-Agent Systems, 2003

447 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/e7dbbbd570006bb6255667f87780a325277ca591

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Control and Systems Engineering

Citations in the corpus

Teachable robots: Understanding human teaching behavior to build more effective robot learners (2008, unknown): “Input Output”
How Humans Teach Agents - A New Experimental Perspective (2012, background): “In past work, many of these agents only exploit [2, 18, 26] and some, especially those that use reinforcement learning, explore or exploit [15, 37, 40].”
Multiagent Collaborative Task Learning through Imitation (2007, methodology): “Our approach utilizes the learning by experienced demonstration technique [18], in which the agent is fully under the expert's control while continuing to experience the task through its own sensors.”
Discovering Motor Programs by Recomposing Demonstrations (2020, background): “Learning from Demonstration: The ﬁeld of learning from demonstrations (LfD) (Nicolescu & Mataric, 2003) has sought to learn to perform tasks from a set of demonstrated behaviors.”
Programming by demonstration: dual-arm manipulation tasks for humanoid robots (2004, background): “In [ 6 ] the authors concentrate on the role of interaction during task learning.”
Socially guided intrinsic motivation for robot learning of motor skills (2013, background): “Reward Reward-like signals ( si H ∈ R ) or “good or bad” indications ( si H ∈ {− 1 ; 1 } ) are common in reinforcement-based approaches, which beneﬁt considerably from the formalism of reinforcement learning (Nicolescu and Mataric 2003; Thomaz 2006; Thomaz and Breazeal 2008).”
Incremental learning of full body motion primitives and their sequencing through human motion observation (2012, background): “In addition, some researchers focusing on learning the relationships between primitives utilize pre-programmed, static primitive behaviors, for example Nicolescu and Matarić (2001, 2003, 2005), Ilg et al. (2004), and Dominey et al. (2008).”
Recent Advances in Robot Learning from Demonstration (2020, background): “Methods that learn task plans from demonstration can either learn a primitive sequence (132, 133, 141, 145) or a primitive hierarchy (128, 146, 147, 148, 149) (see Fig.”
Active choice of teachers, learning strategies and goals for a socially guided intrinsic motivation learner (2012, background): “An interactive learner who not only listens to the teacher, but actively requests for the information it needs and when it needs help, has been shown to be a fundamental aspect of social learning [13, 31, 40].”
Statistical Learning by Imitation of Competing Constraints in Joint Space and Task Space (2009, background): “A large body of work explored the use of a symbolic representation to both the learning and the encoding of skills and tasks, see e.g. [5, 6].”