Physics · Dynamics
EmergentRuleSearch#9c19
Vary the rules of simple agents in simulation to see which yield interesting collective behavior
Mechanism
Start from a vague intuition about a phenomenon, define simple individual agents with tunable rules, and run the simulation while varying parameters. Observe under which conditions the phenomenon (such as a dominance hierarchy) emerges. The result is discovered, not specified, and deep learning is a large instance of the same search over unit rules.
Invariants
- Rules apply to individual agents, not to the outcome
- The outcome is observed after running, not assumed
Failure modes
- Oversimplified world that produces artifacts of the model
- Funding pressure to name the outcome before the search
Identity
The name is the hash of the definition. Change a word and it becomes a different pattern.
- Sema ID
sema:EmergentRuleSearch#mh:SHA-256:9c193dcf9edc7b6c9eadab7da9f27ab778970100b5efc1dacfb5c334bfcf8792- Reference
EmergentRuleSearch#9c19- Ring
- 2
- Tier
- 3