Search papers, labs, and topics across Lattice.
1
0
2
By cleverly mining historical data for invariances, ISD-linUCB achieves superior regret in non-stationary bandits, even when the reward model changes rapidly.
