• Publications
  • Influence
Automatic discovery and transfer of MAXQ hierarchies
TLDR
We present an algorithm, HI-MAT (Hierarchy Induction via Models And Trajectories), that discovers MAXQ task hierarchies by applying dynamic Bayesian network models to a successful trajectory from a source reinforcement learning task. Expand
  • 85
  • 9
  • PDF
Transfer in variable-reward hierarchical reinforcement learning
TLDR
We introduce the problem of Variable-Reward Transfer Learning where the objective is to speed up learning in a new SMDP by transferring experience from previous MDPs that share the same dynamics but have different rewards. Expand
  • 96
  • 5
  • PDF
Automatic Discovery and Transfer of Task Hierarchies in Reinforcement Learning
TLDR
We demonstrate that causally motivated task hierarchies transfer more robustly than other kinds of detailed knowledge that depend on the idiosyncrasies of the source domain. Expand
  • 13
  • 4
  • PDF
Hierarchical structure discovery and transfer in sequential decision problems
TLDR
A new approach to the discovery of hierarchical structure in sequential decision problems by learning simple action models, leveraging these models to analyze non-hierarchically generated trajectories from multiple source problems in a robust causal fashion, and discovering hierarchical structure that transfers to all problems whose actions share the same causal dependencies as those in the source problems. Expand
  • 5
  • 1
Probabilistic Ontology Trees for Belief Tracking in Dialog Systems
TLDR
We introduce a novel approach for robust belief tracking of user intention within a spoken dialog system. Expand
  • 17
  • PDF
Multi-Agent Shared Hierarchy Reinforcement Learning
Hierarchical reinforcement learning facilitates faster learning by structuring the policy space, encouraging reuse of subtasks in different contexts, and enabling more effective state abstraction. InExpand
  • 14
Autonomous Learning of Action Models for Planning
TLDR
This paper introduces two new frameworks for learning action models for planning. Expand
  • 8
  • PDF
Automatic Induction of MAXQ Hierarchies
Scaling up reinforcement learning to large domains requires leveraging the structure in the domain. Hierarchical reinforcement learning has been one of the ways in which the domain structure isExpand
  • 5
  • PDF
Dynamic language modeling using Bayesian networks for spoken dialog systems
TLDR
We introduce a new framework employing statistical language models (SLMs) for spoken dialog systems that facilitates the dynamic update of word probabilities based on dialog history. Expand
  • 4
  • PDF
Efficient Learning of Action Models for Planning
We consider the problem of learning action models for planning in two frameworks and present general sufficient conditions for efficient learning. In the mistake-bounded planning framework, theExpand
  • 2
  • PDF
...
1
2
...