研究テーマ on 阿部 拳之https://bakanaouji.github.io/ja/research/Recent content in 研究テーマ on 阿部 拳之Hugo -- gohugo.ioja© 2026 阿部 拳之Bandits and Online Learninghttps://bakanaouji.github.io/ja/research/bandits-online-learning/Mon, 01 Jan 0001 00:00:00 +0000https://bakanaouji.github.io/ja/research/bandits-online-learning/オンライン環境で意思決定をしながら効率的に学習するには？Fairness in Recommender Systems and Allocationhttps://bakanaouji.github.io/ja/research/fairness-recsys-allocation/Mon, 01 Jan 0001 00:00:00 +0000https://bakanaouji.github.io/ja/research/fairness-recsys-allocation/限られた資源や機会を公平に配分するには？Language Model Alignment and Preference Optimizationhttps://bakanaouji.github.io/ja/research/language-model-alignment/Mon, 01 Jan 0001 00:00:00 +0000https://bakanaouji.github.io/ja/research/language-model-alignment/言語モデルの出力を人間の選好にどう整合させるか？Learning Dynamics and Equilibrium Computation in Gameshttps://bakanaouji.github.io/ja/research/learning-dynamics-equilibrium-games/Mon, 01 Jan 0001 00:00:00 +0000https://bakanaouji.github.io/ja/research/learning-dynamics-equilibrium-games/ナッシュ均衡へ高速に収束する学習アルゴリズムとは？Reinforcement Learning and Sequential Decision Makinghttps://bakanaouji.github.io/ja/research/reinforcement-learning-sequential-decision/Mon, 01 Jan 0001 00:00:00 +0000https://bakanaouji.github.io/ja/research/reinforcement-learning-sequential-decision/逐次的な意思決定において、方策をどう改善・評価するか？