Probabilistic Reasoning

• Daphne Koller, Avi Pfeffer. Probabilistic frame-based systems. AAAI-98.
• Avi Pfeffer, Daphne Koller. Semantics and inference for recursive probability models. AAAI-00.
  These two papers introduce a way to unify frame-based systems and Bayes nets.
  

Machine Learning

• Radford M. Neal. Connectionist learning of belief networks. Artificial intelligence 56 (1992) 71--113.
  This work shows stochastic belief networks link probabilistic reasoning formalisms, such as Bayes networks, and machine learning formalisms, such as the Boltzmann machines.
• Geoffery Hinton, Simon Osindero, Yee-Whye Teh. A fast algorithm for deep belief nets. 2006.
  I understand this work as a new efficient algorithm for learning belief nets above. Experimental result include hand-written digit recognition.
  

Predictive, Psychology and Philosophy

• David B. Fogel. Imagining machines with imagination. IEEE November 1999.
  This is a review of George Morton's paper "machines with imagination". It addresses the importance of randomness in artificial intelligence, and talks about an evolutional view of AI.
• James L. McClelland and Timothy T. Rogers. The parallel distributed processing approach to semantic cognition. April 2003.
  This review paper talks about a neural networks (PDP) model of semantics cognition, which seems to me quite plausible, though my feeling is that the feed-forward network model may be limited.