I think the G-P map alignment you talk about here has been formulated (amongst many other things) in https://doi.org/10.1088/2632-072X/ad9cdc—Biological arrow of time (Prokopenko et al., 2025)
Of course, evolution and SGD have their own idiosyncrasies. Understanding those seems to me as worthwhile as looking for these putative commonalities. When we include evolutionary algorithms, evolution strategies and genetic algorithms for example behave completely different (and noise sensitivity may play a crucial role beyond the zeroth-order approximation!)
Love this.
I think the G-P map alignment you talk about here has been formulated (amongst many other things) in https://doi.org/10.1088/2632-072X/ad9cdc—Biological arrow of time (Prokopenko et al., 2025)
And additivity and averaging should be related to the work on rainbow networks by Menard et al., e.g.: https://arxiv.org/abs/2409.19460
Of course, evolution and SGD have their own idiosyncrasies. Understanding those seems to me as worthwhile as looking for these putative commonalities. When we include evolutionary algorithms, evolution strategies and genetic algorithms for example behave completely different (and noise sensitivity may play a crucial role beyond the zeroth-order approximation!)
https://doi.org/10.1145/3205455.3205474 - ES is more than a finite differences approximator (lehman et al., 2018)
Especially Figure 3 in the Diffusion-Evolution paper https://proceedings.iclr.cc/paper_files/paper/2025/hash/ba5f1233efa77787ff9ec015877dbd1f-Abstract-Conference.html
Would also love to see some papers other people interested in this space enjoy!