首页> 外文期刊>Philosophia >Hitchcock and Sober on Weak Predictivism
【24h】

Hitchcock and Sober on Weak Predictivism

机译:希区柯克和索伯论弱预测主义

获取原文
获取原文并翻译 | 示例
           

摘要

According to Hitchcock and Sober’s argument from overfitting for weak predictivism, the fact that a theory accurately predicts a portion of its data is evidence that it has been formulated by balancing simplicity and goodness-of-fit rather than overfitting data. The core argument consists of two likelihood inequalities. In this paper I show that there is a surprising accommodation-friendly implication in their argument, and contend that it is beset by a substantial difficulty, namely, there is no good reason to think that their second likelihood inequality is true.
机译:根据希区柯克和索伯(Hitchcock and Sober)关于弱预测主义过度拟合的观点,理论可以准确预测其一部分数据这一事实证明,它是通过平衡简单性和拟合优度而不是过度拟合数据而形成的。核心论点包括两个似然不等式。在本文中,我证明了他们的论点中有令人惊讶的适应友好的含义,并认为这受到很大困难的困扰,即没有充分的理由认为他们的第二种可能性不等式是正确的。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号