Evolution as Learning

The Tron system was intended to function as one intelligent, learning opponent to challenge humanity. The strategy of this virtual agent is generated by the random mixture of Tron robots in the evolving population; 18% of the games being played by new, untested agents, exploring new strategy space. The remaining games are played by those agents considered the best so far -- survivors from previous generations, exploiting previous knowledge. In terms of traditional AI, the idea is to utilize the dynamics of evolution by selection of the fittest as a way to create a mixture of experts that create one increasingly robust Tron player.

  

Figure 3.12: Relative strength of the Tron species increases over time, showing artificial learning.

Solving equation (3.9) for all of the computer's games put together yields the performance history of the whole system considered as a single playing entity. Fig. 3.12 shows that the system has been learning throughout the experiment, at the beginning performing at a RS rate below -2.0, and at the end around 0.

But the RS is an arbitrary scale: what does it mean in human terms? The next graph re-scales the RS values in terms of the percent of humans below each value. Beginning as a player in the lower 30 percent, as compared to humans, the Tron system has improved dramatically: by the end of the period it is a top 5% player (fig. 3.13).

  

Figure 3.13: Strength values for the Tron system, plotted as percent of humans below. In the beginning our system performed worse than 70% of all human players. Now it is within the best 5%.