Wednesday, January 9, 2013

Converge: Unifying Science

        In this blog I've hesitated to fully discuss the scope of my personal research because it's too long-winded and disorganized, but every once in a while, a corroborating publication is announced and I find my efforts validated, at least in my own mind.

        I have been cataloging the advance of specific facets of modern science, namely biotechnology, nanotechnology, artificial intelligence, resolution (scaning and playing) technology, and complex systems theory (this includes climatology, sociology, economics etc).  In a way I am resolved as a witness - a historian of the present - in the sense that I have been watching this creature grow according to the principles I've been learning.  There are involved some profound implications of ontology which I have yet to fully explore, and which I will avoid in this article. However, I will outline a few of the guidelines for self-organization. 

        Among them are fractal math, hierarchical hidden Markov chains, non-unitary dynamics, the Constructal Law, the Law of Accelerating Returns; and some of the less important scale-dependent variables, generally involved in the process of evolution.  The principles contained in the framework can be manipulated to produce what is called emergence. The specific and intentional manipulation of that information is what I call intelligent emergence.

        The word evolution owes in part it's etymological existence from Charles Darwin's "on the origin of the species," (1859) a treatise which spawned modern conceptions of biological evolution, namely genetics and its heritable characteristics; in addition to the belief that variations in DNA are random and the selection process is through survival in an environment.

         Since that time, mainstream scientists have been hesitant to acknowledge the mysterious force driving evolution toward that "fitness."  In the course of my study, I have uncovered various redefinitions of evolution; my own personal remix of this word includes quantum physics, atomic physics, chemical and bioligical interaction; as well as all meta-scales up to the brain (there is evidence of cosmological formation based on these principles, but I have not spent enough time on astrophysics to be certain). The simple version of evolution is that it is defined in part by its source, or original values, and its feedback relationship with its environment.  This can be applied to technology rather well, as we have seen the most profound transformation of humanity in only one hundred years (quadrupled the lifespan, exponential creation of information and consumption of energy).  I spoke on the topic of an "emergent hive mind" in 2011 in this post, for those interested in the impact of technology on humanity, and the examination of parallels between super-organisms and humans.


       Evolution occurs on every scale of reality.  Any universe with time will have a form of evolution, although the constraints of how it works will vary depending on the laws of physics.  In our universe, they are such that (constructal law) all systems with flowing energy will attempt to find the most efficient path.   We see this phenomenon in the macroscopic natural world in the form of rivers, trees, the human visual pathway, and in the simplicity of a snowdrift or a sand dune.  This is where we may conceive of the concept of order; which is an often misused (and more often misunderstood) concept in science.

        Order is defined by the success of an architecture in a system.  Why is it that when we boil water, it always occurs at the same temperature?  Why are bubbles shaped as spheres and not cubes, or pyramids? Because the most efficient reorganization of water molecules at 100 degrees C is to "boil." Because bubbles are the most efficient structure for rising energy in boiling water.  The same can be said for the organization of the alveoli in the human lung (or the lungs of any mammal for that matter); which mirror the shapes of river basins and cauliflower. (See self-similarity, fractals). 

       So how is this knowledge useful, beyond these fantastic blog posts?  It is my view that these systematic parameters may be used in order to design metamaterials with emergent properties, as well as provide a rudimentary statistical framework for modeling biocompatible materials integration; chemical drug design; and artificial neural networks (to name only a few).

       But this work is already being done. To my slight chagrin, but not outside my expectation; researchers from a study funded by the Academy of Finland have published "Electrostatic assembly of binary nanoparticle superlattices using protein cages;" taking advantage of the mathematics of complex systems.  These researchers use modeling techniques to predict the outcomes of asymmetries (or symmetries) of various combinations of atoms in a "superlattice." This allows them to predict the electrical, magnetic, optic, and other properties that emerge from their interaction. 

The abstract reads:

   "Here, we show that electrostatically patchy protein cages—cowpea chlorotic mottle virus and ferritin cages—can be used to direct the self-assembly of three-dimensional binary superlattices."

        The implications of this unity between biological modeling and materials science are beyond the applications of any particular material.  This publication (among dozens of others) are symbols for the coming age in science.  Bottom-up tinkering will allow us to treat disease; build materials not currently in existence in the universe; design powerful new drugs; build artificially intelligent neural nets; and countless other applications (literally, the imagination is the limit). Some of the progression of this transition will occur through this process:

1) An understanding of biological design principles, which have been honed over billions of years of iterations, is acquired.
2) the application of those variables to consolidated computational models is made;
3) the integration of biological automatons and small matter physics data; streamlining both design and production;
4) merging of biological application tools and design principles with complex computational models of materials interactions; producing semi-autonomous, novel materials design. 

This is a specific framework for bio-nano materials evolution, however, this behaviour is evident in any system with disparate parts. Consider another model for synthesis:

1) Moore's Law continues to push the boundaries of computational capacity
          A)  Facilitated by the elements listed below in a closed feedback loop.  

2) The core algorithms of human thought are reverse engineered:

                  A)  Neuroscience "cloud" of data is distributed across the internet, creating an ever-shifting web of information;
                  B) breakthroughs in neuroscience are accomplished by vast amounts of expimental data (Blue Brain Project and neural column slices; atomic force microscopy, etc);
                 C) combined with self-organizing mathematical frameworks of "thought," which are task-oriented, and not necessarily biological models, i.e.  a punch can knock someone unconscious, but so can a thick branch.  
                Examples of this include neural nets designed to interpret human language, perceive shapes and objects, and Watson (Watson uses a vast network of diverse algorithms, instead of relying on one unifying mathematical aspect, like what may well be the case in the neocortex (this allows plasticity). 

3)  The algorithms gleaned from computation contribute to its efficiency, further advancing the core mathematics of intelligence.  


4) Intelligence always amplifies everything:
             
A) Intelligent computers facilitate advanced materials science, such as we have seen in carbon nanotubes, graphene, and the more recent bio-nano structures.

 (See Eric Drexler, nanotechnology, for what we might expect from mechanical nanotech in the future).

        What happens when a superintelligent computer discovers new materials to enhance its intelligence, further amplifying it's capacity to make materials, and so on?  It is the topic of some eithical debate over what our relationship to AI will be, and what it will look like. It is a debate in which our children will likely participate.

       This is a vastly oversimplified progression of events, which when extrapolated and applied to different fields, demonstrates a systematic symmetry of progression.  Already, there are consolidated neural, chemical, and biological databases where outcomes can be predicted based on initial conditions, time, and a few simple parameters of evolution.  (See Stephen Wolfram, computing).

        When mastery of these techniques is made, it may be possible for researchers to make high-precision predictions on more than just scientific outcomes; but societal outcomes as well (See Standford Complex Systems Theorists, food prices, rioting).

             In principle, my hypothesis is that these elements of evolution should be applied to our own exploration of evolution itself. That within the progression of atoms to molecules, and they to life, and life to an intellect that can perceive its own components in a meaningful way; there is a self-mirroring inherent at every scale. As science moves along this path, what new reflections will it create? 

       It is my opinion, my not as yet proven statement, that through this new science, we will discover profound truths about our origins and destiny; moving in the spirit of this strange harbinger - toward greater order, diversity, truth, and the beauty that follows.

        It is as Einstein said, "any intelligent fool can build something large, complex and violent; it takes a touch of genius, and a lot of courage, to move in the opposite direction."


Post Edit:

[Since I first wrote this article, a paper published in Nature's Scientific Reports claims that the Universe is shaped like a giant brain.

The Huffington Post reports that "[the] co-author of the study, Dmitri Krioukov from the University of California San Diego, said that while such systems appear very different, they have evolved in very similar ways..For a physicist it's an immediate signal that there is some missing understanding of how nature works," ]

"By performing complex supercomputer simulations of the universe and using a variety of other calculations, researchers have now proven that the causal network representing the large-scale structure of space and time in our accelerating universe is a graph that shows remarkable similarity to many complex networks such as the Internet, social, or even biological networks." - USC news
 

No comments:

Post a Comment