Learning from demonstration in the wild
Learning from demonstration (LfD) is useful in settings where hand-coding behaviour or a reward function is impractical. It has succeeded in a wide range of problems but typically relies on manually generated demonstrations or specially deployed sensors and has not generally been able to leverage th...
| Үндсэн зохиолчид: | , , , , , , , , , , |
|---|---|
| Формат: | Conference item |
| Хэвлэсэн: |
IEEE
2019
|