Shakeel Gavioli-Akilagun’s contribution to the Discussion of ‘the Discussion Meeting on Probabilistic and statistical aspects of machine learning’
Shakeel Gavioli-Akilagun 对机器学习概率与统计讨论会论文的评论
We congratulate the authors for the interesting and thought provoking paper. Lemma 3.2 shows that the generalized likelihood ratio (GLR) test for stability of a linear regression can be viewed as a simple feed forward neural network. However, close inspection of the lemma reveals that the setup rules out several common change point problems. Consider the piecewise polynomial regression: For ξ ’s distributed i.i.d. N(0,1), the likelihood ratio statistic for a change at…