Search papers, labs, and topics across Lattice.
1
0
You can achieve near-optimal regret in multi-armed bandits with heterogeneous noise, even without knowing the best data source *a priori*, by adaptively pruning noisy sources.
