Corpus ID: 14648921

Goal-Based Action Priors

@inproceedings{Abel2015GoalBasedAP,
  title={Goal-Based Action Priors},
  author={David Abel and D. E. Hershkowitz and Gabriel Barth-Maron and Stephen Brawner and Kevin O'Farrell and J. MacGlashan and Stefanie Tellex},
  booktitle={ICAPS},
  year={2015}
}
Robots that interact with people must flexibly respond to requests by planning in stochastic state spaces that are often too large to solve for optimal behavior. In this work, we develop a framework for goal and state dependent action priors that can be used to prune away irrelevant actions based on the robot's current goal, thereby greatly accelerating planning in a variety of complex stochastic environments. Our framework allows these goal-based action priors to be specified by an expert or… Expand
29 Citations
Composable Planning with Attributes
  • 33
  • Highly Influenced
  • PDF
What can I do here? A Theory of Affordances in Reinforcement Learning
  • 11
  • PDF
Learning to Plan in Complex Stochastic Domains
  • PDF
Learning to Plan in Complex Stochastic Domains
  • PDF
Learning Propositional Functions for Planning and Reinforcement Learning
  • 3
  • PDF
Goal Reasoning , Planning , and Acting with A C T O R S I M , The Actor Simulator 1
  • 10
  • Highly Influenced
  • PDF
...
1
2
3
...

References

SHOWING 1-10 OF 39 REFERENCES
Affordances as Transferable Knowledge for Planning Agents
  • 12
  • PDF
Toward Affordance-Aware Planning
  • 6
  • PDF
What good are actions? Accelerating learning using learned action priors
  • 29
  • PDF
Anticipating Human Activities Using Object Affordances for Reactive Robotic Response
  • 288
  • PDF
Between MDPs and Semi-MDPs: A Framework for Temporal Abstraction in Reinforcement Learning
  • 2,408
  • PDF
Genetically evolved macro-actions in AI planning problems
  • 21
  • PDF
An algebraic approach to abstraction in reinforcement learning
  • 83
  • PDF
Macro-FF: Improving AI Planning with Automatically Learned Macro-Operators
  • 209
  • PDF
Knowledge Processing for Autonomous Robot Control
  • M. Tenorth, M. Beetz
  • Computer Science
  • AAAI Spring Symposium: Designing Intelligent Robots
  • 2012
  • 26
  • PDF
Improving Action Selection in MDP's via Knowledge Transfer
  • 86
  • PDF
...
1
2
3
4
...