论文信息 - Conditional Independence in Testing Bayesian Networks

Conditional Independence in Testing Bayesian Networks

Testing Bayesian Networks (TBNs) were introduced recently to represent a set of distributions, one of which is selected based on the given evidence and used for reasoning. TBNs are more expressive than classical Bayesian Networks (BNs): Marginal queries correspond to multi-linear functions in BNs and to piecewise multi-linear functions in TBNs. Moreover, TBN queries are universal approximators, like neural networks. In this paper, we study conditional independence in TBNs, showing that it can be inferred from d-separation as in BNs. We also study the role of TBN expressiveness and independence in dealing with the problem of learning with incomplete models (i.e., ones that miss nodes or edges from the data-generating model). Finally, we illustrate our results on a number of concrete examples, including a case study on Hidden Markov Models.

[1] Surya Ganguli,et al. On the Expressive Power of Deep Neural Networks , 2016, ICML.

[2] Adnan Darwiche,et al. A differential approach to inference in Bayesian networks , 2000, JACM.

[3] Judea Pearl,et al. Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[4] Allan Pinkus,et al. Multilayer Feedforward Networks with a Non-Polynomial Activation Function Can Approximate Any Function , 1991, Neural Networks.

[5] G. Lewicki,et al. Approximation by Superpositions of a Sigmoidal Function , 2003 .

[6] J. Pearl. Comment: Understanding Simpson’s Paradox , 2013, Probabilistic and Causal Inference.

[7] Christian Tjandraatmadja,et al. Bounding and Counting Linear Regions of Deep Neural Networks , 2017, ICML.

[8] Razvan Pascanu,et al. On the Number of Linear Regions of Deep Neural Networks , 2014, NIPS.

[9] Adnan Darwiche,et al. On the Relative Expressiveness of Bayesian and Neural Networks , 2018, PGM.

[10] Razvan Pascanu,et al. On the number of inference regions of deep feed forward networks with piece-wise linear activations , 2013, ICLR.

[11] Kurt Hornik,et al. Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[12] Adnan Darwiche,et al. On Relaxing Determinism in Arithmetic Circuits , 2017, ICML.