This website requires JavaScript.

Pearl Causal Hierarchy on Image Data: Intricacies & Challenges

Matej Ze\v{c}evi\'cMoritz WilligDevendra Singh DhamiKristian Kersting
Dec 2022
摘要
Many researchers have voiced their support towards Pearl's counterfactualtheory of causation as a stepping stone for AI/ML research's ultimate goal ofintelligent systems. As in any other growing subfield, patience seems to be avirtue since significant progress on integrating notions from both fields takestime, yet, major challenges such as the lack of ground truth benchmarks or aunified perspective on classical problems such as computer vision seem tohinder the momentum of the research movement. This present work exemplifies howthe Pearl Causal Hierarchy (PCH) can be understood on image data by providinginsights on several intricacies but also challenges that naturally arise whenapplying key concepts from Pearlian causality to the study of image data.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答