From: jeffrey E. <[email protected]> Sent: Wednesday, May 9, 2018 1:56 AM To: Joscha Bach Subject: Re: now.? or read trivers on genetic conflict <=div> On Tue, May 8, 2018 at=11:44 AM Joscha Bach wrote: There is probabl= an evolutionary algorithm that describes the competition of the neural wi=ings during the formation of our brains, i.e. in the development of every =ndividual. Not sure if there is an equivalent of to male/female sexual sel=ction, and what it would look like! =div>On May 8, 2018, at 8:33 AM, jeffrey E. <[email protected] <mailto:[email protected]> > wrote: <=r class="m_-4945126078866447802Apple-interchange-newline"> trivers genetic competion . male female genes. in the =dult On Tue, May=8, 2018 at 2:52 AM Joscha Bach rote: I am still trying to understand your question. </=iv> - What do you mean by genetic conflict in a single individual? - What do you mean by machine state? Perhap= you refer to a cybernetic regulation paradigm? Valentino Braitenberg sugg=sted to understand Al from a regulation perspective: feedback loops to app=oach a target state (homeostatic equilibrium). The behavior regulation can=be described with payoffs across many sub-behaviors (which Minsky called a=ents and agencies). Whenever you get a situation where the local cost func=ions lead to a Nash equilibrium that is incompatible with the global goals=of the organism, you introduce a higher level regulation that imposes offs=ts on the payoff matrices of the lower level agents. This leads to the for=ation of an internal governance hierarchy. I think that this was=the core idea of Rodney Brooks' subsumption architecture. (Eventua=ly, Brooks' work did not pan out, he started building Roombas and =awn movers and became an Al pessimist.) On May 7, 2018, at 9:17 PM, jeffrey E. <[email protected]</a> wrote: my question relate to the=metabphor of genetic conflict in a single indiv , is there an =quivalent . machine internal competition