REINFORCEMENT LEARNING WITH GOAL-DIRECTED ELIGIBILITY TRACES
2004 ◽
Vol 15
(09)
◽
pp. 1235-1247
◽
Keyword(s):
The eligibility trace is the most important mechanism used so far in reinforcement learning to handle delayed reward. Here, we introduce a new kind of eligibility trace, the goal-directed trace, and show that it results in more reliable learning than the conventional trace. In addition, we also propose a new efficient algorithm for solving the goal-directed reinforcement learning problem.
2007 ◽
Vol 19
(6)
◽
pp. 1468-1502
◽
2020 ◽
Vol 34
(05)
◽
pp. 8878-8885
Keyword(s):
1999 ◽
pp. 320-325
◽
Keyword(s):