Challenge : Where is the Impact of Bayesian Networks in Learning ?
Challenge : Where is the Impact of Bayesian Networks in Learning ?
复制标题
挑战：贝叶斯网络对学习的影响在哪里？
DOI：
--
复制
发表时间：
1997
期刊：
影响因子：
--
通讯作者：
Stuart J. Russell
中科院分区：
文献类型：
--
作者：
N. Friedman;Moises Goldszmidty;David Heckermanz;Stuart J. Russell
贝叶斯网络是概率分布的图形表示。在过去的十年中，这些表示已经成为人工智能中不确定性表示的首选方法。今天，它们在现代专家系统、诊断引擎和决策支持系统中发挥着至关重要的作用。近年来，人们对从数据中学习贝叶斯网络产生了浓厚的兴趣。学习这样的模型是可取的，因为有大量现成的工具可以应用上面描述的学习模型。从业人员还声称，自适应贝叶斯网络作为密度估计、数据分析、模式分类和建模的非参数方法具有自身的优势。我们提到的原因包括：它们的语义清晰度和人类的可理解性，易于获取和合并先验知识，易于与最优决策方法集成，学习模型的因果解释的可能性，以及噪声和缺失数据的自动处理。尽管有这些说法，学习贝叶斯网络的方法还没有像神经网络和隐马尔可夫模型等其他技术在模式和语音识别等应用中所产生的影响。在本文中，我们要求研究界识别和表征贝叶斯网络诱导产生关键差异的领域，并量化导致该差异的因素。除了将挑战形式化之外，我们还确定了研究问题，在我们看来，这些问题的解决方案对于迎接这一挑战至关重要。
Bayesian networks are graphical representations of probability distributions. Over the last decade, these representations have become the method of choice for representation of uncertainly in artiicial intelligence. Today, they play a crucial role in modern expert systems, diagnosis engines, and decision support systems. In recent years, there has been much interest in learning Bayesian networks from data. Learning such models is desirable simply because there is a wide array of oo-the-shelf tools that can apply the learned models as described above. Practitioners also claim that adaptive Bayesian networks have advantages in their own right as a non-parametric method for density estimation, data analysis, pattern classii-cation, and modeling. Among the reasons cited we nd: their semantic clarity and understand-ability by humans, the ease of acquisition and incorporation of prior knowledge, the ease of integration with optimal decision-making methods , the possibility of causal interpretation of learned models, and the automatic handling of noisy and missing data. In spite of these claims, methods that learn Bayesian networks have yet to make the impact that other techniques such as neural networks and hidden Markov models have made in applications such as pattern and speech recognition. In this paper, we challenge the research community to identify and characterize domains where induction of Bayesian networks makes the critical diierence, and to quantify the factors that are responsible for that diierence. In addition to formalizing the challenge, we identify research problems whose solution is, in our view, crucial for meeting this challenge.