From Santa Fe Institute Events Wiki
SFI Sherpa Meeting
June 26 - 27, 2019
Santa Fe Institute
Santa Fe, NM
Over the last several decades new discoveries and theories have emerged to study complex systems - networks of adaptive agents - that promise to be better suited to addressing economics, disease, cyber-security, and medical treatment than erstwhile approaches founded on what we might call the classical engineering axioms.
The purpose of this meeting is to explore and discuss these principles. In a recent paper, David Krakauer presented some candidate new principles. We would like the participants to address these and or some of their own that take us beyond the centralized and classical design thinking that dominates current debate. These principles include:
1. Modifying the reward or selective context in which semi-autonomous agents operate with an eye toward better incentives.
2. Accepting significant component error rates and focus on mechanisms that can average and aggregate these effects to acceptable levels in the collective output
3. Design towards distributions of outcomes and not towards deltas - single optimal outcomes -- pursuing average properties throughout
4. Develop mechanisms for controlling non-linear dynamics and predicting and influencing critical transitions
5. Harness adaptation to allow for continued exploration and exploitation rather than coercing systems into single states that require endless iterations of production.