PRODuctive bandits: Importance Weighting No More
–Neural Information Processing Systems
Prod is a seminal algorithm in full-information online learning, which has been conjectured to be fundamentally sub-optimal for multi-armed bandits.
Neural Information Processing Systems
May-31-2025, 15:21:56 GMT
- Genre:
- Research Report > Experimental Study (0.93)
- Industry:
- Education > Educational Setting > Online (0.48)
- Technology: