Francesco Orabona

Latest

• Minimax Optimal Quantile and Semi-Adversarial Regret via Root-Logarithmic Regularizers