Self-Replicating Neural Programs
This introduces a novel paradigm for evolutionary self-replication in neural programs, potentially advancing autonomous AI systems.
The paper tackled the problem of enabling neural networks to self-replicate their training code through an evolutionary paradigm, resulting in more efficient learning as demonstrated by organisms achieving faster reproductive maturity without explicit guidance.
In this work, a neural network is trained to replicate the code that trains it using only its own output as input. A paradigm for evolutionary self-replication in neural programs is introduced, where program parameters are mutated, and the ability for the program to more efficiently train itself leads to greater reproductive success. This evolutionary paradigm is demonstrated to produce more efficient learning in organisms from a setting without any explicit guidance, solely based on natural selection favoring organisms with faster reproductive maturity.