Clipping Loops for Sample-Efficient Dialogue Policy Optimisation
21613Knowledge Triples
| Source Entity | Relationship | Target Entity |
|---|---|---|
| achieves |
Success Rate
Metric |
|
| consistsOf |
Advantage Clipping
Method |
|
| consistsOf |
Loop Clipping
Method |
|
| evaluatedIn |
Cambridge Restaurant Dialogue System
Material |