We study the subject of universal online learning with non-i.i.d. processes for bounded losses. The notion of an universally consistent learning was defined by Hanneke in an effort to study learning theory under minimal assumptions, where the objective is to obtain low long-run average loss for any target function. We are interested in characterizing processes for which learning is possible and whether there exist learning rules guaranteed to be universally consistent given the only assumption that such learning is possible. The case of unbounded losses is very restrictive, since the learnable processes almost surely visit a finite number of points and as a result, simple memorization is optimistically universal. We focus on the bounded setting and give a complete characterization of the processes admitting strong and weak universal learning. We further show that k-nearest neighbor algorithm (kNN) is not optimistically universal and present a novel variant of 1NN which is optimistically universal for general input and value spaces in both strong and weak setting. This closes all COLT 2021 open problems posed by Hanneke on universal online learning.
翻译:我们研究的是与非i.i.d.d. 过程的封闭性在线学习问题。Hanneke界定了普遍一致学习的概念,以努力在最低假设下研究学习理论,目的是为任何目标功能获得低长期平均损失。我们感兴趣的是,将有可能学习的流程定性为低长期平均损失,以及是否存在普遍一致的学习规则,因为唯一的假设是,这种学习是可能的。无限制损失的情况非常严格,因为可以学习的流程几乎肯定会访问有限的点,因此,简单的记忆是乐观的。我们侧重于受约束的设置,并完整地描述接受强弱普遍学习的过程。我们进一步表明,K-earest 邻居算法(kNN)并不乐观地具有普遍性,我们提出了一个新的1NN值变式,该变式在强弱的环境中对一般投入和价值空间都是乐观的。这关闭了Hanneke在普及在线学习上提出的所有2021年COLT公开问题。