Scheduling Straight-Line Code Using Reinforcement Learning and Rollouts

Authors

Amy McGovern, University of Massachusetts - Amherst
Eliot Moss, University of Massachusetts - Amherst
Andrew G. Barto, University of Massachusetts - Amherst

Publication Date

1999

Abstract

The execution order of a block of computer instructions on a pipelined machine can make a difference in its running time by a factor of two or more. In order to achieve the best possible speed, compilers use heuristic schedulers appropriate to each specific architecture implementation. However, these heuristic schedulers are time-consuming and expensive to build. We present empirical results using both rollouts and reinforcement learning to construct heuristics for scheduling basic blocks. In simulation, the rollout scheduler outperformed a commercial scheduler, and the reinforcement learning scheduler performed almost as well as the commercial scheduler.

Comments

This paper was harvested from CiteSeer

Recommended Citation

McGovern, Amy; Moss, Eliot; and Barto, Andrew G., "Scheduling Straight-Line Code Using Reinforcement Learning and Rollouts" (1999). Computer Science Department Faculty Publication Series. 14.
Retrieved from https://scholarworks.umass.edu/cs_faculty_pubs/14