| Summary: | We extend the agent-based models for knowledge diffusion in networks, restricted to <i>random mindless interactions</i> and to <i>“frozen” (static) networks</i>, in order to take into account <i>intelligent agents</i> and <i>network co-evolution</i>. Intelligent agents make decisions under <i>bounded rationality.</i> This is the key distinction of intelligent interacting agents compared to <i>mindless colliding molecules</i>, involved in the usual diffusion mechanism resulting from <i>accidental</i> collisions. The <i>co-evolution</i> of link weights and knowledge levels is modeled <i>at the local microscopic level of “agent-to-agent” interaction</i>. Our network co-evolution model is actually a “<i>learning mechanism”</i>, where weight updates depend on the previous values of <i>both</i> weights and knowledge levels. The goal of our work is to explore the impact of (a) the <i>intelligence</i> of the agents, modeled by the selection-decision rule for knowledge acquisition, (b) the <i>innovation rate</i> of the agents, (c) the <i>number of “top innovators”</i> and (d) the <i>network size</i>. We find that <i>rational intelligent</i> agents <i>transform</i> the network into a <i>“centralized world”</i>, <i>reducing the entropy of their selections-decisions</i> for knowledge acquisition. In addition, we find that the <i>average knowledge</i>, as well as the <i>“knowledge inequality”</i>, grow exponentially.
|
|---|