6 papers · 1 filter
Hellinger loss function for Generative Adversarial Networks
Giovanni Saraceno, Anand N. Vidyashankar, Claudio Agostinelli
We propose Hellinger-type loss functions for training Generative Adversarial Networks (GANs), motivated by the boundedness, symmetry, and robustness properties of the Hellinger dis…
Divergence-Minimization for Latent-Structure Models: Monotone Operators, Contraction Guarantees, and Robust Inference
Lei Li, Anand N. Vidyashankar
We develop a divergence-minimization (DM) framework for robust and efficient inference in latent-mixture models. By optimizing a residual-adjusted divergence, the DM approach recov…
Minimum Hellinger Distance Estimators for Complex Survey Designs
David Kepplinger, Anand N. Vidyashankar
Reliable inference from complex survey samples can be derailed by outliers and high-leverage observations induced by unequal inclusion probabilities and calibration. We develop a m…
Sharp Large Deviations and Gibbs Conditioning for Threshold Models in Portfolio Credit Risk
Fengnan Deng, Anand N. Vidyashankar, Jeffrey F. Collamore
We obtain sharp large deviation estimates for exceedance probabilities in dependent triangular array threshold models with a diverging number of latent factors. The prefactors quan…
Ancestral Inference and Learning for Branching Processes in Random Environments
Xiaoran Jiang, Anand N. Vidyashankar
Ancestral inference for branching processes in random environments involves determining the ancestor distribution parameters using the population sizes of descendant generations. I…
Private Minimum Hellinger Distance Estimation via Hellinger Distance Differential Privacy
Fengnan Deng, Anand N. Vidyashankar
Objective functions based on Hellinger distance yield robust and efficient estimators of model parameters. Motivated by privacy and regulatory requirements encountered in contempor…