EFTA00813239.pdf

From: Joscha Bach To: Jeffrey Epstein <[email protected]> Subject: Re: Date: Tue, 08 May 2018 06:52:55 +0000 I am still trying to understand your question. - What do you mean by genetic conflict in a single individual? - What do you mean by machine state? Perhaps you refer to a cybernetic regulation paradigm? Valentino Braitenberg suggested to understand AI from a regulation perspective: feedback loops to approach a target state (homeostatic equilibrium). The behavior regulation can be described with payoffs across many sub-behaviors (which Minsky called agents and agencies). Whenever you get a situation where the local cost functions lead to a Nash equilibrium that is incompatible with the global goals of the organism, you introduce a higher level regulation that imposes offsets on the payoff matrices of the lower level agents. This leads to the formation of an internal governance hierarchy. I think that this was the core idea of Rodney Brooks' subsumption architecture. (Eventually, Brooks' work did not pan out, he started building Roombas and lawn movers and became an Al pessimist.) On May 7, 2018, at 9:17 PM, jeffrey E. <[email protected]> wrote: my question relate to the metabphor of genetic conflict in a single indiv , is there an equivalent machine internal competition On Mon, May 7, 2018 at 8:06 PM Joscha Bach < > wrote: I think that the whole game is internal. All our pleasure and pain are caused by satisfying or frustrating our needs, which are (roughly speaking) implemented in our reptilian brain. Our neocortex is modeling these needs as purposes, so we can regulate our behavior for satisfying our needs. For example, the physical pain, and hunger and thirst exist on the level of needs, and survival is a purpose, a kind of modeling abstraction, as are food and drink. Many of our purposes are outside of the organism, such as relationship purposes, and social purposes. Even our aesthetic purposes (like Jeffrey enjoying an intellectual standoff between Bannon and Summers) are not really related to the organism. While the needs are somewhat hardcoded, our reaction to them is not (we can modulate pain and pleasure experiences, we can route our regulation around satisfying our needs, and we can repurpose our needs for competence, dominance, affiliation, and aesthetics for almost arbitrary goals. By changing the regulation goals (which is always internal to our mind) we change the payoff matrix. On May 7, 2018, at 8:43 AM, jeffrey E. <[email protected]> wrote: i posit that thre is an internal game with its own payoff matrix , separate from the external . is there a machine internal state equivalent. EFTA00813239  please note The information contained in this communication is confidential, may be attorney-client privileged, may constitute inside information, and is intended only for the use of the addressee. It is the property of JEE Unauthorized use, disclosure or copying of this communication or any part thereof is strictly prohibited and may be unlawful. If you have received this communication in error, please notify us immediately by return e-mail or by e-mail to [email protected], and destroy this communication and all copies thereof, including all attachments. copyright -all rights reserved please note The information contained in this communication is confidential, may be attorney-client privileged, may constitute inside information, and is intended only for the use of the addressee. It is the property of JEE Unauthorized use, disclosure or copying of this communication or any part thereof is strictly prohibited and may be unlawful. If you have received this communication in error, please notify us immediately by return e-mail or by e-mail to [email protected], and destroy this communication and all copies thereof, including all attachments. copyright -all rights reserved EFTA00813240