Wednesday, August 23, 2006

The Singularity

The world and universe are flooded with patterns. Patterns are hidden all around us, from the lowest level of reality to the utmost outer reaches of space. Patterns are also right in front of our faces. Our minds are only equipped to handle certain types of problems and patterns...mainly problems and patterns where we can encode the pattern/problem and come up with an internal representation of it that seems to match with the observed phenomenon in the real world. In many cases there could exist and I believe does exist significant patterns that our minds are not equipped to handle, or perhaps we can sense the pattern, but we can not grasp the pattern's full meaning and implications. For example, we can function on a day to day basis without understanding how our own minds work on every level. We can also function without understanding all of the billions of patterns that exist in the biological world around us, as well as the billions and billions of patterns in the abiotic world/universe as well.

But this disparity between the existence of patterns and our ability to fully grasp their meaning may be shrinking. At the moment the disparity is still very large in some cases (e.g. what makes things alive as opposed to being just a random collection of molecules). It seems entirely possible that we may be reaching a huge watershed in the way we view systems that appear to be irreducibly complicated...at least to our minds. I am of the belief that even the most complicated system can be made elementary if approached from the correct perspective and with the appropriate contextual framework, as well as with the appropriate resources. This is because it appears that most complex dynamic systems end up with some sort of emergent phenomenon that is produced from the underlying set of interactions, and if one can understand the relations between the interactions and the emergent phenomenon, then one could quite possibly break the system down into an easily digestible framework without having to understand every causality or specific relationship. One of the huge disadvantages of the current state of our brains is that, while incredibly adaptable and amazing in their own right, they still are fundamentally very limited in the way in which they can both modify their own architecture as well as the physical limitations of the speed at which neurons fire, and the information density that can be retained per volume of gray matter. Plus, our minds never evolved to deal with decomposing complex dynamic systems that may have many hidden patterns between the underlying set of interactions and any emergent phenomenon. Our minds evolved for many reasons, and understanding complex relationships does appear to be one of the abilities we attained...but only certain types of relationships and abstractions. Other types (such as a system of 1000 coupled ODE's with high degrees of non-linearity) are not as tractable to our minds but could be very useful when describing a neuronal circuit itself, or the complicated signaling transduction pathway between a set of cells.

But I believe that we are reaching a point where we are developing the tools to begin to understand the subtle patterns and relationships that have been hidden from us so far. The incredibly exciting product of this increased understanding is that it will build on itself in a nonlinear fashion...or in other words it will have a snow-balling effect. As our ability to understand and model the subtle and complicated pattern that are all around us increases, the rate at which we understand these systems will also increase. I believe this will occur because of the universality of some of these patterns. For example, if we could fully understand how a gene network would behave for every possible perturbation to its structure, then we would probably be able to extend this knowledge to neuronal networks, signaling transduction pathways, and all other sorts of systems where connectivity between elements and interactions between elements is isomorphic to that of a gene network. Another reason why this understanding will increase exponentially is because new tools and techniques in different and related fields will feed off each other. A good example of this is systems biology, where data and models from many different levels of information of an organism are integrated to produce a more holistic picture of the organism. An example would be looking at how the variation in a phenotype of an organism for some complex trait (e.g. how long a disease lasts) was related to genetic variation in a population (from the genome), the gene expression data (transcriptome), what proteins had the highest concentrations in what cells (proteome), as well as what metabolic pathways were activated (metabolome), and variation in the immune response (e.g. what kinds of immune cells were activated, and what salient structural features about them were correlated with how long the disease lasted). All of these different levels of information use different techniques (e.g. genotyping, microarrays, 2D-SDS Page, protein assays, NMR perhaps...), but when synthesized into an entire model with appropriate considerations for the assumptions that are necessary, it produces a better and better picture of the organism as a whole. This can be said for any complex dynamical system, where the techniques for observing information about the system are improving for many different aspects of the system all the time, as well as our ability to model the increasing quantity of information, and place it in the appropriate context.

What does this all mean in the long run? Well it could mean some pretty amazing things. It could also mean some pretty horrible things. It all depends on how we as a society and world decide to use this knowledge. Here's what I imagine could possibly happen:

As the connectivity and types of interactions evolve over time between models of reality (e.g. the models between gene networks, social networks, neuronal networks, etc...) the way in which we as human beings understand not only reality but ourselves will go through amazing changes. Just like the changes it went through after the theory of evolution, the theory of gravity, the theory of special and general relativity, etc... Except this time these changes will be to an even larger degree, because we will be driving right towards that point that we have never been able to look directly at before...the idea or reason behind what makes life live and what lets us go around having thoughts recognizing that they are thoughts. As we get closer and closer to these points our ability to not only understand but change ourselves on a much more fundamental level will become possible. This is an exciting but scary time. There will be a period of time where we will be on the edge of a precipice, and the slightest gust of wind could send us into an abyss of darkness so bleak it chills me just thinking about it. Yet if we can survive that incredibly precarious period, the possibilities are so amazing I can't even begin to describe them. One image that comes to mind is that of all symbols and internal models that describe reality that are contained in every mind in the world coalescing into a singular symbol which describes not only reality but itself as well. Sort of like a super-consciousness.

These are possibilities and concerns that we need to keep in mind in the coming years as we make what only appears to be incremental progress towards these amazing and frightening events. Because they will happen faster than we could ever believe.

Wednesday, August 02, 2006

Dreaming of New ?Dreams?

Every thought and emotion can be assigned/is a symbol. Symbols affect one another in often mysterious ways. Many symbols can not be expressed adequately with language, because every time one tries to capture the essence of the symbol, the meaning diffuses away, leaving an empty shell of a thought or idea. This happens to me with the most meaningful symbols. I find that communicating a symbol that describes the why of what I'm interested in and what I want to do on a very basic level is impossible. Instead I'm stuck with the bulky and often unwieldy apparatus of language, whose inherent impetus towards obfuscating meaning gets in the way of the ideas I would like to share with others.

Language hides meaning because it imposes constraints that are merely a subset of the possible constraints that are working in my brain. This is an important distinction to make. Whereas there are many many constraints in my brain, keeping it from either falling apart physically or mentally, the number of possible constraints in the mind is much greater than those provided for by the mere syntactical structure of language. There are only about 171,000 words that are in current use according to the Oxford English Dictionary. This is compared to the billions of neurons and trillions of connections that exist in the human brain. Hence the inherent complexity that is possible in the mind is much greater than what can be adequately conveyed with words...at least for now. It is true that one can string many words together to try to capture the essence of a single thought, but in many cases this fails to do justice to the thought or symbol, whose inherent structure in one's mind is so crystal clear. And there is also a much larger set of unknowns which is even more mysterious. The following diagram illustrates these relationships:
This diagram shows how the constraints on the set of possible thoughts are organized into categories. The unknown constraints rule overall, with reality, my mind, and language all being subsets of the largest possible set of constraints.

But of course nothing is ever this simple...sometimes language breaks out of the system and tries to define itself. A computer program is merely a language that is implemented with some sort of processing unit that records operations and instructions and takes input and gives output. Some could argue that language is then inherent in our own brains, because each neuron is talking in it's own language to other neurons (like little processors), and the total sum of their interactions produces a new higher level language (there are probably many levels involved here) that is conscious thought. But then language then is able to start talking about itself, since a conscious mind can come up with ideas like language...so where does language come from? Is language then all there is to reality? Or is there something more fundamental going on here? And what about the unknowns?

I have no idea what the answer to these questions are, or if they are even meaningful questions. I feel like maybe I have succumbed to the very weakness I spoke of earlier...that of not being able to adequately convey the meaning of an idea...yet when that idea is "not being able to convey the meaning of idea" things start to get very tangled and confused. And this is where language starts to get very confusing...whenever it starts referring to itself, language does some pretty bizarre things that our minds can not handle. We can of course "jump out of the system" for many of these silly paradoxes (e.g. "this statement is a lie") but in some cases we can't jump out of the system and we are stuck in some bizarre loop, going round and round and round and never knowing it, or knowing that we are even in the loop.

It's kind of like being stuck in a dream within a dream. Even if you do wake up there will always be some other dream you are stuck in...