A probabilistic framework for representing and visually recognizing complex multi-agent action is presented. Motivated by work in model-based object recognition and designed for the recognition of action from visual evidence , the representation has three components: (1) temporal structure descriptions representing the logical and temporal relationships between agent goals, (2) belief networks ...

A probabilistic framework for representing and visually recognizing complex multi-agent action is presented. Motivated by work in model-based object recognition and designed for the recognition of action from visual evidence, the representation has three components: (1) temporal structure descriptions representing the temporal relationships between agent goals, (2) belief networks for probabili...

We define and investigate a structure incorporating what is true, what is known and what is believed by a rational agent in models of the S4.2 logic of knowledge. The notion of KBR-structures introduced, provides a fine-grained modal analysis of an agent’s epistemic state, actually one that differentiates knowledge from belief and accounts for an agent without full introspective power (concerni...

Convolutional neural net-like structures arise from training an unstructured deep belief network (DBN) using structured simulation data of 2-D Ising Models at criticality. The convolutional structure arises not just because such a structure is optimal for the task, but also because the belief network automatically engages in block renormalization procedures to “rescale” or “encode” the input, a...

We address the issue of belief revision in a multi-agent setting. We represent agents' beliefs in a semantic manner, through a Kripke structure, and model a communication process by which agents communicate their beliefs to one another. We define a revision rule that can be applied even when agents have contradictory beliefs. We study its properties and show that agents need not agree after com...

We address the issue of belief revision in a multi-agent setting. We represent agents’ beliefs in a semantic manner, through a Kripke structure, and model a communication process by which agents communicate their beliefs to one another. We define a revision rule that can be applied even when agents have contradictory beliefs. We study its properties and show that agents need not agree after com...