Amari 对信息几何和机器学习的贡献是众所周知的。在这里,,我们重新审视 Amari 在贝叶斯对偶性方面的工作,该工作尚未受到太多关注。我们将 Amari的 贝叶斯对偶性与 Bayes 规则的凸对偶性联系起来。使用这种连接,,我们提出了 Amari的 贝叶斯对偶性的概括,并讨论了它与现代人工智能的相关性。
Amari的 contributions to information geometry and machine learning are well known. Here, we revisit Amari的 work on Bayesian duality which has not received as much attention. We connect Amari的 Bayesian duality to a convex duality of Bayes rule. Using this connection, we present a generalization of Amari的 Bayesian duality and discuss its relevance for modern artificial intelligence.
科目: 人工智能 (cs.AI); 机器学习 (cs.LG); 机器学习 (stat.ML)
Subjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Machine Learning (stat.ML)