Search Results for author:
Found 2 papers, 0 papers with code
Marginal Weighted Maximum Log-likelihood for Efficient Learning of Perturb-and-Map models
We consider the structured-output prediction problem through probabilistic approaches and generalize the "perturb-and-MAP" framework to more challenging weighted Hamming losses, which are crucial in applications.
Parameter Learning for Log-supermodular Distributions
Then, to learn parameters, given that our approximation of the log-partition function is an expectation (over our own randomization), we use a stochastic subgradient technique to maximize a lower-bound on the log-likelihood.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
