Distributional Reinforcement Learning with Regularized Wasserstein Loss Ke Sun
–Neural Information Processing Systems
Empirically, we show that SinkhornDRL consistently outperforms or matches existing algorithms on the Atari games suite and particularly stands out in the multi-dimensional reward setting.
Neural Information Processing Systems
Nov-19-2025, 17:06:46 GMT
- Country:
- North America > Canada
- Quebec > Montreal (0.04)
- Alberta > Census Division No. 11
- Edmonton Metropolitan Region > Edmonton (0.04)
- Asia > China
- Heilongjiang Province > Harbin (0.04)
- North America > Canada
- Genre:
- Research Report
- Experimental Study (1.00)
- New Finding (0.67)
- Research Report
- Industry:
- Leisure & Entertainment > Games > Computer Games (0.56)
- Technology: