Automatic Programming of Behavior-Based Robots Using Reinforcement Learning

By Sridhar Mahadevan, J. Connell. 1991

How researchers see this paper

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

Citations in the corpus

Reinforcement learning in robotics: A survey (2013, background): “One of the earliest papers in robot reinforcement learning (Mahadevan and Connell, 1992) introduced this idea by statistically clustering states to speed up a box-pushing task with a mobile robot, see Figure 1a.”
Reinforcement Learning: A Survey (1996, neutral): “Mahadevan and Connell (1991b) used the dual approach: they (cid:12)xed the gating function, and supplied reinforcement functions for the individual behaviors, which were learned.”
Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning (1999, background): “For example, a robot may have pre-designed controllers for servoing joints to positions, picking up objects, and visual search, but still face a difficult problem of how to coordinate and switch between these behaviors [17,32,35,39,40,43,61,79].”
Using communication to reduce locality in distributed multiagent learning (1997, unknown): “In robot learning, Mahadevan & Connell (1991) focused on state gener-alization and controller modularity applied to a sonar-based robot learningto push a box in an experiment complementary our our rst one, which15 addresses the same problem with a similar approach, but introduces com-munication to…”
Multi Agent Reinforcement Learning Independent vs Cooperative Agents (2003, methodology): “In this paper, I use reinforcement learning to study intelligent a g e n ts (Mahadevan & Connel 1991, Lin 1991, Tan 1991).”
Real-time reinforcement learning by sequential Actor-Critics and experience replay (2009, methodology): “This idea of repeating many similar operations to the same event, called experience replay (Mahadevan & Connell, 1992; Lin, 1992; Lin & Mitchell, 1992; Cichosz, 1999), was popular a few years ago but has received little attention recently.”
An approach to learning mobile robot navigation (1995, unknown): “Recently, Q-learning and other related reinforcement learning and dynamic programming algorithms [2,34] have been applied successfully to robot manipulation [ 8], robot control [ 16 ] and games [35].”
THE ROLE OF EXPLORATION IN LEARNING CONTROL (1992, background): “Consequently, semi-uniform distributed exploration performs exploitation with probability Pbest, and random exploration otherwise [9, 29, 30, 60].”
Generalization in Reinforcement Learning: Safely Approximating the Value Function (1994, methodology): “This approach, which dates back to uses of Legendre polynomials in DP [Bellman et al., 19631, has recently worked well on several dynamic control problems [Mahadevan and Connell, 1990, Lin, 1993] and succeeded spectacularly on the game of backgammon [Tesauro, 1992, Boyan, 1992].”