That's the nature of assumptions isn't it: Some camps think one algorithm is sufficient for General Intelligence, Some camps think multiple algorithms are necessary for General Intelligence. The only way to prove which assumption is correct is to produce a working system and evaluate its capabilities. Good luck Boris.
From: [email protected] To: [email protected] Subject: Re: [agi] Attention to abstractions Date: Thu, 7 Mar 2013 11:48:22 -0500 > It appears you have everything figured out. Congratuations. I don't, but it's hard to talk about your problems before explaining your solutions. > However, sometimes people have their own definitions for words that may differ from yours. I find your taxonomy hugely redundant. You are trying to include everything & everyone in your model, & end up with a huge mess. Just a personal observation, if you don't mind, - I think you are too nice :). Making a discovery usually involves realizing that everyone else, up to now, is full of shit. From: Piaget Modeler Sent: Thursday, March 07, 2013 11:29 AM To: AGI Subject: RE: [agi] Attention to abstractions Hi Boris, It appears you have everything figured out. Congratuations. However, sometimes people have their own definitions for words that may differ from yours. Take for example, Consolidation. In PAM-P2, consolidation is exactly automaticity + forgetting. That is the definition and implementation. We seek to find useful chains of activated concepts and combine them while deleting useless concepts. Also, in PAM-P2 we have the concept of a Serendipity and also the concept of a Shock. A serendipity in problem solving is arriving at a goal state before the complete execution of your plan. Some factor in the world allowed you to get to your goal earlier than expected. This is not magic, this is a new feature of the environment which must be internally synthesized, and understood. Causes for the serendipity must be ascribed and tested so that the conditions under which it happens can be accommodated. In a similar fashion, a shock is an unexpected and undesirable side effect resulting from executing a plan. Some factor in the world was not accounted for in the plan. This factor is also a new feature of the environment which must be internally synthezied and understood. Likewise, causes for the shock must be ascribed and tested so that the conditions under which it happens can be accommodated. Just something for you to consider. Date: Thu, 7 Mar 2013 09:24:38 -0600 Subject: Re: Re: [agi] Attention to abstractions From: [email protected] To: [email protected] "Serendipity" is another word for magic. Any search is uncertain, by definition, but your chances are much better when you're searching in a relevant areas. Subconscious searching only works when you build-up enough of relevant representations, which is what conscious focus is for. No, it's a word for stumbling into the right answer by accident. No magic or anything else mysterious involved. It happens most often when you think you know what is relevant but have left something important out without realizing it. yours [bottleneck] is inductive generalization If you're going to tell me I have a cognitive shortcoming, please back it up with whatever reason you believe so, and be precise about what you think that shortcoming is. On Thu, Mar 7, 2013 at 8:12 AM, Boris Kazachenko <[email protected]> wrote: Aaron, > And what of serendipity? "Serendipity" is another word for magic. Any search is uncertain, by definition, but your chances are much better when you're searching in a relevant areas. Subconscious searching only works when you build-up enough of relevant representations, which is what conscious focus is for. > I have also read of several recent studies that show that performance actually improves after distractions and/or breaks. Breaks, definitely. Distractions may work like breaks in a short term, but will keep distracting you in a long term. The simple fact is that you have a fixed amount of resources between the ears. > A healthy, happy mind is capable of much more than one constantly under stress, and being asocial is definitely stressful for most people, as is being in pain or focusing for too long without a rest period. You get used to it. Like I said, breaks & naps are great, but socializing is just as likely to be frustrating as relaxing. Depending on what you expect from it. Positive emotions from socializing can be replaced by those from work, - that's conditioning. It's hard at first, but so is anything worthwhile. > It's not just about how much time you spend thinking about the task at hand; the quality of thought during that time is also important. Quality follows quantity, as long as you're focusing on a right level of generalization. > Being overly obsessive is probably not in your best interest. I am obsessive long-term, quite relaxed short-term. We probably have very different bottlenecks here: mine is deductive focus, yours is inductive generalization. And the latter must come first, - it defines what a problem is. So, I may have a leg up :). On 03/06/13, Aaron Hosford<[email protected]> wrote: And what of serendipity? I find sometimes my best ideas hit me out of nowhere when I happen to observe something seemingly unrelated that I then tie back to the problem I'm trying to solve. A random stimulus sometimes causes ideas to bump up against each other which otherwise never would have. I have also read of several recent studies that show that performance actually improves after distractions and/or breaks. When I was in high school, I participated for several years in a Number Sense competition, which consisted of taking a 10 minute high-speed test in which math problems had to be solved without showing any work or correcting any answers. Skipping or missing problems resulted in negative points, so the goal was to get as far through the 80 problems as possible without missing or skipping any. Each school year I would practice daily, gradually improving my score. Then, I would relax over the summer break. When I returned to school again the next fall, I would find my score had jumped up far above what it would have been had I continued to improve at the same rate that I did while I was practicing daily. I suspect the resting time allowed the concepts to go "offline" temporarily and be reorganized in my head, which could not be done as effectively while I had to keep them ready for active use. Most people would just call it getting a fresh perspective. Don't underestimate the value of self-maintenance, either. A healthy, happy mind is capable of much more than one constantly under stress, and being asocial is definitely stressful for most people, as is being in pain or focusing for too long without a rest period. It's not just about how much time you spend thinking about the task at hand; the quality of thought during that time is also important. I don't believe the human brain is designed for constant focus. It seems more likely, given our evolutionary history, that it is designed for spurts of high-intensity focus with frequent rests in between. So you might find your average productivity rate increases when you allow yourself to alternate between focusing and resting, so long as resting isn't excessive. Being overly obsessive is probably not in your best interest. AGI | Archives | Modify Your Subscription AGI | Archives | Modify Your Subscription ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
