Work

Optimal Combination Policies for Error Exponent Maximization in Social Learning

FWCI: 1.288
Cites: 21
Cited by: 1
Related to: 10