鹿島・山田研から以下の5本の論文が機械学習の難関国際会議NeurIPS 2019に採択されました。

• Ryoma Sato, Makoto Yamada, Hisashi Kashima.
  Approximation Ratios of Graph Neural Networks for Combinatorial Problems.
  Advances in Neural Information Processing Systems (NeurIPS 2019).
  #Graph Neural Network と Distributed Local Algorithm の関係性を理論的に示し、より強力なGNNを提案.
• Yasutoshi Ida, Yasuhiro Fujiwara, Hisashi Kashima.
  Fast Sparse Group Lasso.
  Advances in Neural Information Processing Systems (NeurIPS 2019).
  # 枝刈りによるGroup Lassoの高速化
• Tam Le, Makoto Yamada, Kenji Fukumizu, Marco Cuturi.
  Tree-Sliced Variants of Wasserstein Distances.
  Advances in Neural Information Processing Systems (NeurIPS 2019).
  #木構造データ間のWasserstein距離を高速に求める方法を提案.
• Jenning Lim, Makoto Yamada, Bernhard Schoelkopf, Wittawat Jitkrittum
  Kernel Stein Tests for Multiple Model Comparison.
  Advances in Neural Information Processing Systems (NeurIPS 2019).
  #Selective Inferenceを用いたGoodness-of-fitテスト
• Rafael Pinot, Laurent Meunier, Alexandre Araujo, Hisashi Kashima, Florian Yger, Cédric Gouy-Pailler, Jamal Atif.
  Theoretical Evidence for Adversarial Robustness Through Randomization.
  Advances in Neural Information Processing Systems (NeurIPS 2019).
  # モデルへの攻撃に対するランダム化による対策についての理論的考察