Recent Submissions

  • Policy Shaping: Integrating Human Feedback with Reinforcement Learning 

    Griffith, Shane; Subramanian, Kaushik; Scholz, Jonathan; Isbell, Charles L.; Thomaz, Andrea L. (Georgia Institute of TechnologyNeural Information Processing System, 2013)
    A long term goal of Interactive Reinforcement Learning is to incorporate non- expert human feedback to solve complex tasks. Some state-of -the-art methods have approached this problem by mapping human information to ...
  • Multimodal Real-Time Contingency Detection for HRI 

    Chu, Vivian; Bullard, Kalesha; Thomaz, Andrea L. (Georgia Institute of TechnologyInstitute of Electrical and Electronics Engineers, 2014-09)
    Our goal is to develop robots that naturally engage people in social exchanges. In this paper, we focus on the problem of recognizing that a person is responsive to a robot’s request for interaction. Inspired by human ...
  • Generating Human-like Motion for Robots 

    Gielniak, Michael J.; Liu, C. Karen; Thomaz, Andrea L. (Georgia Institute of TechnologySage Publications, 2013-07)
    Action prediction and fluidity are key elements of human-robot teamwork. If a robot’s actions are hard to understand, it can impede fluid HRI. Our goal is to improve the clarity of robot motion by making it more humanlike. ...
  • Object Focused Q-Learning for Autonomous Agents 

    Cobo, Luis C.; Isbell, Charles L., Jr.; Thomaz, Andrea L. (Georgia Institute of TechnologyACM Press, 2013)
    We present Object Focused Q-learning (OF-Q), a novel reinforcement learning algorithm that can offer exponential speed-ups over classic Q-learning on domains composed of independent objects. An OF-Q agent treats the state ...
  • Effective robot task learning by focusing on task-relevant objects 

    Lee, Kyu Hwa; Lee, Jinhan; Thomaz, Andrea L.; Bobick, Aaron F. (Georgia Institute of TechnologyInstitute of Electrical and Electronics Engineers, 2009-10)
    In a robot learning from demonstration framework involving environments with many objects, one of the key problems is to decide which objects are relevant to a given task. In this paper, we analyze this problem and propose ...
  • Social Learning Mechanisms for Robots 

    Thomaz, Andrea L.; Cakmak, Maya (Georgia Institute of Technology, 2009)
    There is currently a surge of interest in service robotics—a desire to have robots leave the labs and factory floors to help solve critical issues facing our society, ranging from eldercare to education. A critical issue ...
  • Controlling Social Dynamics with a Parametrized Model of Floor Regulation 

    Chao, Crystal; Thomaz, Andrea L. (Georgia Institute of TechnologyBrigham Young University, 2012)
    Turn-taking is ubiquitous in human communication, yet turn-taking between humans and robots continues to be stilted and awkward for human users. The goal of our work is to build autonomous robot controllers for successfully ...
  • Automatic Task Decomposition and State Abstraction from Demonstration 

    Cobo, Luis C.; Isbell, Charles L., Jr.; Thomaz, Andrea L. (Georgia Institute of TechnologyInternational Foundation for Autonomous Agents and Multiagent Systems, 2012-06)
    Both Learning from Demonstration (LfD) and Reinforcement Learning (RL) are popular approaches for building decision-making agents. LfD applies supervised learning to a set of human demonstrations to infer and imitate the ...
  • Trajectories and Keyframes for Kinesthetic Teaching: A Human-Robot Interaction Perspective 

    Akgun, Baris; Cakmak, Maya; Yoo, Jae Wook; Thomaz, Andrea L. (Georgia Institute of Technology, 2012-03)
    Kinesthetic teaching is an approach to providing demonstrations to a robot in Learning from Demonstration whereby a human physically guides a robot to perform a skill. In the common usage of kinesthetic teaching, the robot's ...
  • Timing in Multimodal Turn-Taking Interactions: Control and Analysis Using Timed Petri Nets 

    Chao, Crystal; Thomaz, Andrea L. (Georgia Institute of TechnologyBrigham Young University, 2012)
    Turn-taking interactions with humans are multimodal and reciprocal in nature. In addition, the timing of actions is of great importance, as it influences both social and task strategies. To enable the precise control and ...
  • Keyframe-based Learning from Demonstration Method and Evaluation 

    Akgun, Baris; Cakmak, Maya; Jiang, Karl; Thomaz, Andrea L. (Georgia Institute of TechnologySpringer, 2012-06)
    We present a framework for learning skills from novel types of demonstrations that have been shown to be desirable from a human-robot interaction perspective. Our approach –Keyframe-based Learning from Demonstration (KLfD)– ...
  • Enhancing Interaction Through Exaggerated Motion Synthesis 

    Gielniak, Michael J.; Thomaz, Andrea L. (Georgia Institute of Technology, 2012-03)
    Other than eye gaze and referential gestures (e.g. pointing), the relationship between robot motion and observer attention is not well understood. We explore this relationship to achieve social goals, such as influencing ...
  • Multi-Cue Contingency Detection 

    Lee, Jinhan; Chao, Crystal; Bobick, Aaron F.; Thomaz, Andrea L. (Georgia Institute of TechnologySpringer, 2012-04)
    The ability to detect a human's contingent response is an essential skill for a social robot attempting to engage new interaction partners or maintain ongoing turn-taking interactions. Prior work on contingency detection ...
  • Anticipation in Robot Motion 

    Gielniak, Michael J.; Thomaz, Andrea L. (Georgia Institute of TechnologyInstitute of Electrical & Electronics Engineers, 2011)
    Robots that display anticipatory motion provide their human partners with greater time to respond in interactive tasks because human partners are aware of robot intent earlier. We create anticipatory motion autonomously ...
  • Task-Aware Variations in Robot Motion 

    Gielniak, Michael J.; Liu, C. Karen; Thomaz, Andrea L. (Georgia Institute of TechnologyInstitute of Electrical & Electronics Engineers, 2011-05)
    Social robots can benefit from motion variance because non-repetitive gestures will be more natural and intuitive for human partners. We introduce a new approach for synthesizing variance, both with and without constraints, ...
  • Towards Grounding Concepts for Transfer in Goal Learning from Demonstration 

    Chao, Crystal; Cakmak, Maya; Thomaz, Andrea L. (Georgia Institute of TechnologyInstitute of Electrical & Electronics Engineers, 2011-08)
    We aim to build robots that frame the task learning problem as goal inference so that they are natural to teach and meet people's expectations for a learning partner. The focus of this work is the scenario of a social robot ...
  • Human-like Action Segmentation for Option Learning 

    Shim, Jaeeun; Thomaz, Andrea L. (Georgia Institute of TechnologyInstitute of Electrical & Electronics Engineers, 2011)
    Robots learning interactively with a human partner has several open questions, one of which is increasing the efficiency of learning. One approach to this problem in the Reinforcement Learning domain is to use options, ...
  • Simon plays Simon says: The timing of turn-taking in an imitation game 

    Chao, Crystal; Lee, Jinhan; Begum, Momotaz; Thomaz, Andrea L. (Georgia Institute of TechnologyInstitute of Electrical & Electronics Engineers, 2011)
    Turn-taking is fundamental to the way humans engage in information exchange, but robots currently lack the turn-taking skills required for natural communication. In order to bring effective turn-taking to robots, we must ...
  • Combining function approximation, human teachers, and training regimens for real-world RL 

    Zang, Peng; Irani, Arya; Zhou, Peng; Isbell, Charles L.; Thomaz, Andrea L. (Georgia Institute of TechnologyInternational Foundation for Autonomous Agents and Multiagent Systems, 2010)
  • Exploiting social partners in robot learning 

    Cakmak, Maya; DePalma, Nick; Arriaga, Rosa; Thomaz, Andrea L. (Georgia Institute of TechnologySpringer, 2010)
    Social learning in robotics has largely focused on imitation learning. Here we take a broader view and are interested in the multifaceted ways that a social partner can in uence the learning process. We implement four ...

View more