Programmable Reinforcement Learning Agents

@inproceedings{Andre2000ProgrammableRL,
  title={Programmable Reinforcement Learning Agents},
  author={David Andre and Stuart J. Russell},
  booktitle={NIPS},
  year={2000}
}
We present an expressive agent design language for reinforc ement learning that allows the user to constrain the policies considere by the learning process.The language includes standard features such a par meterized subroutines, temporary interrupts, aborts, and memor y variables, but also allows forunspecifiedchoices in the agent program. For learning that which isn’t specified, we present provably convergent l arning algorithms. We demonstrate by example that agent programs writt en n the… CONTINUE READING
Highly Influential
This paper has highly influenced 26 other papers. REVIEW HIGHLY INFLUENTIAL CITATIONS
Highly Cited
This paper has 132 citations. REVIEW CITATIONS

5 Figures & Tables

Topics

Statistics

01020'02'04'06'08'10'12'14'16'18
Citations per Year

133 Citations

Semantic Scholar estimates that this publication has 133 citations based on the available data.

See our FAQ for additional information.