If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#211
|
|||
|
|||
Low carb diets
On Sun, 21 Dec 2003 21:35:47 GMT, "OmegaZero2003"
wrote: "Wayne S. Hill" wrote in message ... OmegaZero2003 wrote: BTW, as a PS to my other answer post, here are some linear systems. - those characterizable by linear algebra. there are lots of these! 'ang on there, when we refer to linearity in such systems, we're referring to linearity of the dynamics, i.e., x-dot = A * x + b 'ang on there yourself!!! Do you know WHAT characterizes a linear system mathematically? Do you know the difference between an additive versus a multiplicative factor? Well there is your answer! In case you have not understood - will will take it slowly at first. Atend: here is help for you: "The response of a [ linear ] system to a sum of inputs is the sum of the responses to each individual input separately. All engineers should know what superposition is. Are you trying to be condescending? These two nice properties allow a whole range of tools to be applicable in designing linear systems and predicting their behavior. Some more examples of linear systems in real life: a) Frequency filters -- circuits which only pass low frequencies and reject high, or vice-versa. They only work within a narrowly-bounded input-output range. Too much signal input and they will break (non-linear). b) Delays are linear. Echos from faraway canyons are linear. Shout twice as loud, get an echo twice as loud. Two people shouting at the same time comes back as two people echoing at the same time. Two much volume and you will heat the air, changing the transmission characteristics (speed) of your pressure wave (non-linear). c) Many different kinds of economic systems -- looking at the apple juice production (output) vs. the apple crop yield, for example. This too will fail linearity at the extremes of yield. A bumper crop one year could mean an inability to harvest all the fruit before it goes bad. d) Limiting cases of non-linear systems for small inputs: Even if the system's response may not satisfy the equation above exactly, it often will well enough for small enough inputs. In this case, even if the number of apples bought by consumers, say, is inversely proportional to the price of apples, you can still model small changes around a reference price with linear systems (but beware when the inputs get large!). (http://van.hep.uiuc.edu/van/qa/secti...20020319233837. htm) Note that the last item (d) talks about linearization/signal decomposition/approximation_techniques of/for non-linear systems, which I am NOT talking about. We are talking about bona fide LINEAR systems fulfilling all the mathematical/theoretical properties that decades of science has determined qualifies a system as "Linear" *by definition*!!!! Real systems are always limited on their inputs. Real systems break down at some point if too much energy enters the system. Linearity is only a trait of a system over its useful operating range. I am also not talking about trivial examples of what is a linear system in toto (including extreme), but is subclassed to non-interesting behavior (like an amp NOT in saturation mode). In reality, people who make things that work must take nonlinearity into account. It's the only way to make reliable systems. You may not be thinking in toto, but I am, because I have to. We know what linear is. We also know it doesn't really exist. We try to make things work as linear as possible, because many useful things exhibit a limited version of it (like amps and filters). The definition of linearity for that amp doesn't allow for limiting the input. If it were truly linear, you wouldn't have to place extra limits on it. So I think you are coming around to accept the assertion that linear systems don't really exist. I will admit, though, they are nice on paper. This omits simple algebraic behaviors (because x-dot = 0). - Hamiltonian oscillators and like systems. (the direction field specifically) Not of great interest in biological systems, except as a backdrop. We were not restricting the discussion to biological systems. My original retort to the OP's statement: "All systems are non-linear" was to say: That is not true. It is easy to falsify hypotheses of the form: "All...(x) are (y)" in the sense that only one counter-example need be provided. FYI, a special class of linear control systems known as singularly perturbed control systems, uses the Hamiltonian approach ( recursive) approach based on t exact pure-slow and pure-fast decoupling of optimal control problems. Another interesting special class of linear systems:: "Linear systems with non-rational transfer functions (In this project linear systems described by partial differential equations having non-rational transfer functions are studied. The aim of this project is to analyse the dynamic behaviour and properties of linear input-output systems with non-rational transfer functions (such as flexible robot arms and heat processes) in the frequensy domain. For some classes of such systems it is possible to develop the overall transfer function in analytical form by the transmission matrix method. The interaction between different parts of the system (including the way they are coupled to each other) can then be analysed. The transmission matrix method has been succesfully applied to multi-link flexible robot arms and to buckling of multi-segment columns. Another application concerns the stability of heat processes described by parabolic partial differential equations. Based on the transmission matrix method a Nyquist stability test was developed for sandwich-layered materials with linear inner heat source." (M. Vajta DISC Project 1999) Good grief. Robot arms are the last thing in the world that would be linear. There may be some use of linear mathematics in the control law, but that's it. Linear math in control algorithms is the norm. for a linear system characterized using N first-order linear homogeneous differential equations with constant coefficient can be found at: (with a little more detailed math): http://www.mathpages.com/home/kmath440/kmath440.htm I leave it to the reader to come up with only three biological systems that are so charaterized (this is an easy quiz to see if you understand what a linear system is). You can also look up sparse linear systems for funzies! - continuous-time systems like electrical networks, many mechanical systems Aside from the fact that linearity is an approximation in all such systems, they all have their nonlinear limits. The nonlinearity is whatever it is that keeps the systems operating in their linear range. - any discrete system with a transfer function whose input, response and output functions depend on one variable Huh? I think you have to learn the meaning of linear. Note that I am NOT talking about linearisation of non-linear systems (which seems to be talked about in this thread instead of the major contentions about what exists.)!!! I am not talking about signal decomposition to acheive a linear treatment (what you called piecewise). Here is some help. Note that he points out some examples in the text. "Signals, Linear Systems, and Convolution Professor David Heeger Characterizing the complete input-output properties of a system by exhaustive measurement is usually impossible. Instead, we must find some way of making a finite number of measurements that allow us to infer how the system will respond to other inputs that we have not yet measured. We can only do this for certain kinds of systems with certain properties. If we have the right kind of system, we can save a lot of time and energy by using the appropriate theory about the system's responsiveness. Linear systems theory is a good time-saving theory for linear systems which obey certain rules. Not all systems are linear, but many important ones are. When a system qualifies as a linear system, it is possible to use the responses to a small set of inputs to predict the response to any possible input. This can save the scientist enormous amounts of work, and makes it possible to characterize the system completely." Please note the statement: "Not all systems are linear, but many important ones are." And now for a little math: "Linear Systems A system or transform maps an input signal x(t) into an output signal y(t): y(t) = T[x(t)]; where T denotes the transform, a function from input signals to output signals. Systems come in a wide variety of types. One important class is known as linear systems. To see whether a system is linear, we need to test whether it obeys certain rules that all linear systems obey. The two basic tests of linearity are homogeneity and additivity. 4 Homogeneity. As we increase the strength of the input to a linear system, say we double it, then we predict that the output function will also be doubled. For example, if the current injected to a passive neural membrane is doubled, the resulting membrane potential fluctuations will double as well. This is called the scalar rule or sometimes the homogeneity of linear systems. Additivity. Suppose we we measure how the membrane potential fluctuates over time in response to a complicated time-series of injected current x1(t). Next, we present a second (different) complicated time-series x2(t). The second stimulus also generates fluctuations in the membrane potential which we measure and write down. Then, we present the sum of the two currents x1(t) + x2(t) and see what happens. Since the system is linear, the measured membrane potential fluctuations will be just the sum of the fluctuations to each of the two currents presented separately. Superposition. Systems that satisfy both homogeneity and additivity are considered to be linear systems. These two rules, taken together, are often referred to as the principle of superposition. Mathematically, the principle of superposition is expressed as: T( x1 + x2) = T(x1) + T(x2) (2) Homogeneity is a special case in which one of the signals is absent. Additivity is a special case in which = = 1. Shift-invariance. Suppose that we inject a pulse of current and measure the membrane potential fluctuations. Then we stimulate again with a similar pulse at a different point in time, and again we measure the membrane potential fluctuations. If we haven't damaged the membrane with the first impulse then we should expect that the response to the second pulse will be the same as the response to the first pulse. The only difference between them will be that the second pulse has occurred later in time, that is, it is shifted in time. When the responses to the identical stimulus presented shifted in time are the same, except for the corresponding shift in time, then we have a special kind of linear system called a shift-invariant linear system. Just as not all systems are linear, not all linear systems are shift-invariant. In mathematical language, a system T is shift-invariant if and only if: y(t) = T[x(t)] implies y(t " That is pretty much what I told the OP. Not ALL systems are non-linear. Period, End of story. QED and all that. Gabel, Robert A. and Roberts, Richard A., 1973, Signals and Linear Systems, New York: John Wiley & Sons, 415 pp. Gaskill, Jack D., 1978, Linear Systems, Fourier Transforms, and Optics, New York: John Wiley & Sons, 554 pp. Lathi, B. P., 1992, Linear Systems and Signals, Carmichael, California: Berkeley-Cambridge Press, 656 pp. Lewis, Laurel J., Reynolds, Donald K., Bergseth, F. Robert, Alexandro, Jr., Frank J., 1969, Linear Systems Analysis, New York: McGraw-Hill Book Company, 489 pp. Schwarz, Ralph J. and Friedland, Bernard, 1965, Linear Systems, New York: McGraw-Hill Book Company, 521 pp The most common form of discrete systems is the iteration (strobing based on time or state) of a continuous system. A single-variable system would be of a form x-dot = f(x) Here, f can be (generally is) a nonlinear function of x, so the system will show nonlinear behavior, both continuously and discretely. - any systems preserving homogeneity (output proportional to input) and superposition (a way of combining linear functions such that the result is a linear function) In other words, linear systems. And there are lots of those! Do yourself a favor and merely google "linear system", read what you care to, then come back. I do not believe you do not get this. I can give you a good start: Linear System Theory and Design by Chi-Tsong Chen Read Chapter 2! Even non-deterministic systems can be modeled using statistics for linear dynamics. Not in their dynamics. You do not know what you are talking about. This is the classic engineer's mistake of characterizing in statistical terms what is not understood. When you delve into the NLD of such systems, you gain true insight into what makes them work. I can give examples. I have more examples in my head than you can provide. I worked on this for 20 years. But the main point to not be belabored is that there are linear systems on nature and manmade (1) Most systems *are* non-linear but some of those are characterizable using linear methods to some degree of accuracy; you did make something like this point. That is an approximation that people sometimes find useful to make. That doesn't make it so, especially when you take the system outside of the limited context in which you placed it for convenience. There are systems that are approximated as linear; that is called *linearization* and there is a gamut of math to deal with how to do that properly! But there are many systems that are inherently and demonstrably linear. -- -Wayne --- Proton Soup "If I drink water I will have to go to the bathroom and how can I use the bathroom when my people are in bondage?" -Saddam Hussein |
#212
|
|||
|
|||
Low carb diets
"Proton Soup" wrote in message ... On Sun, 21 Dec 2003 21:35:47 GMT, "OmegaZero2003" wrote: "Wayne S. Hill" wrote in message ... OmegaZero2003 wrote: BTW, as a PS to my other answer post, here are some linear systems. - those characterizable by linear algebra. there are lots of these! 'ang on there, when we refer to linearity in such systems, we're referring to linearity of the dynamics, i.e., x-dot = A * x + b 'ang on there yourself!!! Do you know WHAT characterizes a linear system mathematically? Do you know the difference between an additive versus a multiplicative factor? Well there is your answer! In case you have not understood - will will take it slowly at first. Atend: here is help for you: "The response of a [ linear ] system to a sum of inputs is the sum of the responses to each individual input separately. All engineers should know what superposition is. Are you trying to be condescending? Who? Me? Sorry. I was probably a little. It seemed elementary considering the responses seemed to be intelligent but missed major aspects of this. |
#213
|
|||
|
|||
Low carb diets
OmegaZero2003 wrote:
"Proton Soup" wrote... All engineers should know what superposition is. Are you trying to be condescending? Who? Me? Sorry. I was probably a little. It seemed elementary considering the responses seemed to be intelligent but missed major aspects of this. No they didn't. That was a representation in your own wetware. -- -Wayne |
#214
|
|||
|
|||
Low carb diets
OmegaZero2003 wrote:
Did you bother to see the papers on computing and representation I referenced? Nope! 8-) See, I have a limited capacity for curiosity, and I suspect we're arguing about subtleties that verge on the meaningless. At the very least, a lot is being lost in translation. I do not care what the neurophsyiological response of a crayfish CNS is; there ain't any instruments yet that can tell just what processes and properties in *any* part of a NN exactly *represent* a "piece" of information - even given what the definition of "information" is (beyond a difference). The problem I have with this is that there's a point, with a very small number of neurons, where this distinction vanishes. This is the same problem most researchers have with the idea of "emergence". Beyond a high level of complexity, the network representation is probably closest to a high- dimensional hologram. At a low level of complexity, it's probably pretty much like an ANN. Is the distinction one of topological significance, or is it really all the same thing? In terms of computational neuroscience and ANNs(artificial neural net), remember that the ANN is a couple orders of magnitude less sophisticated (at least) (using simple I/O transforms and connection schemes used to build multi-layer ANNS) than the real thing in situ. I never claimed otherwise: I brought ANN's into the discussion because of the clear understanding of the roles of threshold and saturation on their behavior. -- -Wayne |
#215
|
|||
|
|||
Low carb diets
OmegaZero2003 wrote: "Proton Soup" wrote in message ... On Sun, 21 Dec 2003 06:29:45 GMT, "OmegaZero2003" wrote: BTW, as a PS to my other answer post, here are some linear systems. - those characterizable by linear algebra. there are lots of these! - Hamiltonian oscillators and like systems. (the direction field specifically) - continuous-time systems like electrical networks, many mechanical systems Only simple RLC electrical networks fall into this category. And even then, it's just a theoretical assumption over the useful operating range. Too much current or voltage or flux will flux up your circuit. Linear electrical networks only exist on paper. My original point to the OP on the topic was a retort to the statement that *all* systems are nonlinear. That is not true. - any discrete system with a transfer function whose input, response and output functions depend on one variable - any systems preserving homogeneity (output proportional to input) and superposition (a way of combining linear functions such that the result is a linear function) Even non-deterministic systems can be modeled using statistics for linear dynamics. But the main point to not be belabored is that there are linear systems on nature and manmade (1) Most systems *are* non-linear but some of those are characterizable using linear methods to some degree of accuracy; you did make something like this point. (1) Schwarz, Ralph J. and Friedland, Bernard, 1965, Linear Systems, New York: McGraw-Hill Book Company, 521 pp. --- Proton Soup "If I drink water I will have to go to the bathroom and how can I use the bathroom when my people are in bondage?" -Saddam Hussein As several people have pointed out, linear vs. non--linear, per se, is meaningless. It all depends on the description. Some things are exactly linear (in the right description.) Sometimes linear is just local -- but even if local, this can provide quantitative information. (The example that comes to mind is the tumbling book. Two directions are stable, one is not. Linear analysis shows this.) -- Tom Morley | Same roads | Same rights | Same rules AIM: DocTDM |
#216
|
|||
|
|||
Low carb diets
In article ,
Elzinator no one@com wrote: Lyle McDonald wrote: Ok, since the book project this was originally written for is unlikely to ever get done, I figured I'd post it. It's a long (11 pages) chapter/piece examining the pros/cons of the major dietary camps Dude, just upload it to your website! That's what it's for. I had snagged it, fixed typos and formatting, and put in in the archives at: http://www.trygve.com/mfwalylediet.html I haven't put in any links to it, but I can if that's okay. -- soc.singles FAQ [ Nyx Net, free ISP ] Misc.Fitness.Weights page www.trygve.com/ssfaq.html [ http://www.nyx.net ] www.trygve.com/mfw.html today's special featu Santa Claus, Fugitive From Justice on America's Most Wanton: http://www.trygve.com/mostwanton.html |
#217
|
|||
|
|||
Low carb diets
"Tom Morley" wrote in message link.net... OmegaZero2003 wrote: "Proton Soup" wrote in message ... On Sun, 21 Dec 2003 06:29:45 GMT, "OmegaZero2003" wrote: BTW, as a PS to my other answer post, here are some linear systems. - those characterizable by linear algebra. there are lots of these! - Hamiltonian oscillators and like systems. (the direction field specifically) - continuous-time systems like electrical networks, many mechanical systems Only simple RLC electrical networks fall into this category. And even then, it's just a theoretical assumption over the useful operating range. Too much current or voltage or flux will flux up your circuit. Linear electrical networks only exist on paper. My original point to the OP on the topic was a retort to the statement that *all* systems are nonlinear. That is not true. - any discrete system with a transfer function whose input, response and output functions depend on one variable - any systems preserving homogeneity (output proportional to input) and superposition (a way of combining linear functions such that the result is a linear function) Even non-deterministic systems can be modeled using statistics for linear dynamics. But the main point to not be belabored is that there are linear systems on nature and manmade (1) Most systems *are* non-linear but some of those are characterizable using linear methods to some degree of accuracy; you did make something like this point. (1) Schwarz, Ralph J. and Friedland, Bernard, 1965, Linear Systems, New York: McGraw-Hill Book Company, 521 pp. --- Proton Soup "If I drink water I will have to go to the bathroom and how can I use the bathroom when my people are in bondage?" -Saddam Hussein As several people have pointed out, linear vs. non--linear, per se, is meaningless. It all depends on the description. Some things are exactly linear (in the right description.). Sometimes linear is just local -- but even if local, this can provide quantitative information. Sure - but that is known as a subclassed system. You are only examining it within a range of I/O/Xfer_function(s). Which is fine as far as it goes, I was referring to a system that exhibits true linear behavior throughout all known or extraploated ranges of input. (The example that comes to mind is the tumbling book. Two directions are stable, one is not. Linear analysis shows this.) Yes; although the point that was made , as I now understand, was about biological system, which certainly have more than the bulk of examples of non-lineear systems. -- Tom Morley | Same roads | Same rights | Same rules AIM: DocTDM |
#218
|
|||
|
|||
Low carb diets
"Wayne S. Hill" wrote in message ... OmegaZero2003 wrote: "Proton Soup" wrote... All engineers should know what superposition is. Are you trying to be condescending? Who? Me? Sorry. I was probably a little. It seemed elementary considering the responses seemed to be intelligent but missed major aspects of this. No they didn't. That was a representation in your own wetware. How do you know I am not an ANN! You still have not shown how robustness is a function of saturation and/or thresholding. Although after thinking about it, in an oblique way, one can make up a story about it - a mind game ala Einstein. I.e., theoretically, I can imagine that a system can be thought of as robust if it escapes deterioration/degradation and/or elimination from the context/environment if it exhibits saturation/thresholding and that prevents state spaces leading to elimination. Coming up with a *real* example of such a charaterization in nature is left to the reader. -- -Wayne |
#219
|
|||
|
|||
Low carb diets
"Wayne S. Hill" wrote in message ... OmegaZero2003 wrote: Did you bother to see the papers on computing and representation I referenced? Nope! 8-) See, I have a limited capacity for curiosity, and I suspect we're arguing about subtleties that verge on the meaningless. At the very least, a lot is being lost in translation. OK. I do not care what the neurophsyiological response of a crayfish CNS is; there ain't any instruments yet that can tell just what processes and properties in *any* part of a NN exactly *represent* a "piece" of information - even given what the definition of "information" is (beyond a difference). The problem I have with this is that there's a point, with a very small number of neurons, where this distinction vanishes. First, I suspect that that type of system is uninteresting. Second, it probably does not exist as a real system in situ' one can take away only so much of a system auntil it ceases *being* anything like what you were trying to show in the first place. Such is the case with a system whose function is representation (and transformation/translation/signalling). Third, that distinction is not a quatitative one - it is qualitative. take away the neurochemical soup for example, and what you show about information representation is apt to be misleading at best. *Analysis* (in the form of reductionism)is not always a good approach when dealing with complexity . This is the same problem most researchers have with the idea of "emergence". Beyond a high level of complexity, the network representation is probably closest to a high- But that is only the network representation; one of several maps, none of which is the territory. And most neuroscience researchers or AI researchers for that matter, so not have a problem with emergence. It is quite well described and accepted. Again, a really good book is Alwyn Scott's! I recommend it to any scientist I speak with (just as I recommend Wolfram's work, and Bucky Fuller's Synergetics). dimensional hologram. At a low level of complexity, it's probably pretty much like an ANN. Is the distinction one of topological significance, or is it really all the same thing? I don't follow; I don;t think even a highly-connected network like the brain has properties at the network level (nodes, connections, vertices etc.) that are appropriate in a discussion about holographical metaphors. Now, quantum effects, or other field effects -now we're talking. In terms of computational neuroscience and ANNs(artificial neural net), remember that the ANN is a couple orders of magnitude less sophisticated (at least) (using simple I/O transforms and connection schemes used to build multi-layer ANNS) than the real thing in situ. I never claimed otherwise: I brought ANN's into the discussion because of the clear understanding of the roles of threshold and saturation on their behavior. YEs - OK - I understand. But those are man-made systems that, llike I said above, are so abstracted, or rahter, simplified from the actual NN in the CNS -in situ with all the attendent functions provided by messenger molecules, densities , field effects etc., that the analysis has analysed away any chance of getting a parsimonious, satisfying sanswer to things like representation in a real brain. -- -Wayne |
#220
|
|||
|
|||
Low carb diets
In article , Wayne S. Hill
wrote: Heh: we're definitely talking past each other. And to about six other groups we don't need traffic with. Can you trim your headers along with your carbs please? |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Atkins diet may reduce seizures in children with epilepsy | Diarmid Logan | General Discussion | 23 | December 14th, 2003 11:39 AM |
Low Carb Week in Review | Dave N | General Discussion | 0 | November 24th, 2003 12:06 AM |
Latest Low Carb News | Dave N | General Discussion | 1 | November 18th, 2003 07:13 AM |
What is low carb? | Jarkat2002 | General Discussion | 7 | October 30th, 2003 02:21 PM |
named vs. homegrown diets Curiosity about posters who drop out of this NG | JayJay | General Discussion | 16 | September 27th, 2003 02:16 AM |