Reinforcement learning for sensing strategies

By Cody C. T. Kwok, Dieter Fox. 2004

78 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications, Mechanical Engineering

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

1. Sergey Levine
2. Jan Peters
3. Pieter Abbeel
4. Jens Kober
5. J. Bagnell
6. Chelsea Finn
7. Gerhard Neumann
8. Vladlen Koltun

In Computer Networks and Communications

1. Eduardo Alonso
2. Jan Peters
3. Jens Kober
4. Michael Fairbank
5. J. Bagnell
6. Kris M. Kitani

In Mechanical Engineering

1. Eduardo Alonso
2. Jan Peters
3. Jens Kober
4. Michael Fairbank
5. J. Bagnell

Citations in the corpus

• Reinforcement learning in robotics: A survey (2013, methodology): “Kwok and Fox (2004) used a mixed discrete and continuous representation of the state-space to learn active sensing strategies in a RoboCup scenario.”
• Closed-Loop Learning of Visual Control Policies (2011, unknown): “More recently, Kwok and Fox (2004) have demonstrated the applicability of RL to learning sensing strategies using Aibo robots.”
• Learning reliable and efficient navigation with a humanoid (2010, background): “Kwok and Fox [14] apply reinforcement learning to increase the performance of soccerplaying robots by active sensing.”
• Value-gradient learning (2012, methodology): “Variants of these methods have produced successes in control problems [6], [7], but they can be very slow since Bellman’s condition needs meeting over the entire continuous state space for optimality.”