Full Professor, The University of Tokyo
1 paper at NeurIPS 2025
We present surrogate regret upper bounds for online structured prediction with bandit and delayed feedback.