We analyze the use of built-in policies, or macro-actions, as a form of domain knowledge
that can improve the speed and scaling of reinforcement learning algorithms. Such macro …

We present part of a new algorithm for automatically growing macro-actions online in a
reinforcement learning framework. We call this algorithm acQuire-macros. We present …

Several researchers have proposed reinforcement learning methods that obtain advantages
in learning by using temporally extended actions, or macro-actions, but none has carefully …

Temporally extended actions (eg, macro actions) have proven very useful for speeding up
learning, ensuring robustness and building prior knowledge into AI systems. The options …

Reinforcement learning addresses the problem of learning to select actions in order to
maximize one's performance in unknown environments. To scale reinforcement learning to …

We present PolicyBlocks, an algorithm by which a reinforcement learning agent can extract
useful macro-actions from a set of related tasks. The agent creates macroactions by finding …

We present a method for automatically constructing macro-actions from scratch from
primitive actions during the reinforcement learning process. The overall idea is to reinforce …

This paper discusses why traditional reinforcement learning methods, and algorithms
applied to those models, result in poor performance in situated domains characterized by …

To address the shortcomings of reinforcement learning (RL) a number of researchers have
focused recently on ways to take advantage of structure in RL problems and on ways to …

A common problem in reinforcement learning is finding a balance between exploration
(attempting to discover new features about the world by a selecting sub-optimal action) and …