ChrisChunick.com

[context][unknown] or [context]x

What comes after the known context? Traditionally information is gathered/inputted from left to right, what comes after information, is unknown, and the previous known information determines what comes next. It looks like “[context]x” where x is an unknown bit, and is used in modern AI, compression, and information theory in general.

Back in 2004, while going to school for my Network Administrator diploma, I coined a term “Spatial Decomposition” to describe my goal of achieving the removal and addition of information in a non-linear fashion with a 2 dimensional space of information. I have a working decomposer that uses a simple zero context (input) for determining the probability of an event occurring. Alone this predictor has no knowledge of much of anything, it accumulates two core values at a minimum. It can produce multiple probabilities using different techniques, as well as mix probabilities including weighted to determine the unknown. On it’s own it would be similar to one artificial neuron in a neural net where their is still a lot of room for the transformation of the signal with other ‘neurons’. The difference from modern inputs, is that the input of information can come from the left side and the right side narrowing the guess by a large amount as well a neuron on it’s own is very ineffective at making a guess. One Spatial Decomposition ‘neuron’ can produce significant results. This particular ‘neuron’ on it’s own with no context has no real memory requirements, just needing computational power.

What if we could determine the unknown using the both the left and right hand side instead of the traditional linear left context for input?

Wouldn’t it make it much easier to determine say the following –

“How [unknown] you today?

Instead of

“How [unknown]”

Is this possible? Yes it is, using Spatial Decomposition, where the context of information is no longer required to be isolated preceding a symbol.

The principal equation of a layer of Spatial Decomposition / Recomposition is fairly simple:

SpatialBounds+SpatialLayers=SpatialLayers-SpatialLayer;

Where a layer is not required to be known, it’s essentially free.

note: This was written with the assistance and collaboration with multiple AI’s, including following my direction, information I requested and helping to identify what I described to the AI. Visit an AI detector https://gptzero.me/.

In today’s interconnected world, the phenomenon of identity dilution is worth examining closely. As global culture becomes increasingly homogenized, the unique traits that define individual and cultural identities are at risk of being overshadowed. This dilution is evident in various aspects of modern life, resulting in the erosion of unique self-expression and creativity.

With the constant influence of social media and technology, it’s easy to see how individual identities start to blur. People often conform to dominant trends and societal expectations, and in doing so, they lose a bit of their unique self-expression. The pressure to fit in can stifle our individuality, leading to a more homogenized society.

I’ve seen people struggle to differentiate their true selves from the personas they feel they need to project. This confusion about personal values and beliefs often diminishes self-esteem and personal satisfaction, making people feel inadequate when they can’t stand out. The stress and anxiety from trying to align with societal expectations can make life feel pretty inauthentic.

The impact of this homogenization is particularly concerning when it comes to innovation. Unique perspectives and innovative ideas are often suppressed in environments that prioritize conformity, stifling creativity and progress. Those who resist conforming might feel alienated or marginalized, leading to a sense of disconnection from the larger community.

There’s also a phenomenon referred to as “social mimicry.” The trend towards extreme individuality can paradoxically lead to a new kind of conformity. As people strive to express their uniqueness, they often adopt similar patterns of behavior, dress, and thought, creating a homogenized culture of so-called individualism. This dilutes true individuality and stifles the richness of diverse perspectives, which can further hinder innovation and lead to more identity confusion and social alienation.

My greater concern lies in its similar impact on innovation and technological progress, constraining the possibilities for creative ingenuity. Addressing these core issues of identity dilution seems crucial to nurturing a rich and varied intellectual landscape, especially for those of us who are deeply invested in advanced technology.

Someone once tried to tell me their are no absolutes in the universe when it came to beliefs in regards to philosophy and or religion. I corrected them, without a long explanation, the answer to that question is mathematics. The one universal language in helping understand almost everything.

One of the most fundamentally overlooked pieces of mathematics is non mutual exclusivity or in general something that is not mutually exclusive. This holds some basis in quantum mechanics, artificial intelligence, compression, information theory, as well as almost everything people encounter in daily life but just don’t know how to resolve.

The best non-mathematical definition that I’ve come across that suits the above for me, would be:

“If two things are not mutually exclusive, it means the existence and occurrence of one does not necessarily mean the other cannot coexist.”

Co-Pilot had this to say about my short article:

In the midst of writing my semi-neural net compressor I have introduced a scaling function based off the premise of an Activation function for a neural net – stretch and squash. Because I always work within a normalized domain of 0.0 to 1.0 I was looking for a logarithmic type function that would work similar to a sigmoid to bring my probabilities closer to a best guess or at least a better prediction.

Below is a graph showing a hermite basis function of a Cubic Hermit Spline , which acts as a sigmoid in the domain of 0.0 to 1.0. The function is very simple and results are phenomenal. In comparison I show a linear curve, of course with no scaling and a hyperbolic tangent that obviously is skewed to fit within the domain 0.0 to 1.0. I have also tested other functions such as log sigmoid and double exponential sigmoid with decent results. This simple function results in a logarithmic curve that seems to produce amazing results. Of course outside the domain I use, the function is definitely a fake sigmoid.

The formula is very basic: p(x) = -2x^3+3x^2

* ^ = exponent

Activation Functions

I’ve been developing compression techniques for 17 years now and have many unpublished items. One great thing about compression is it never gets old. Data never changes and Moores law does not apply to the field of digital compression. Data compression is bound by speed, memory, compression ratio very similar to the business triangle of quality, speed, and price.

My most recent soon to be published experimental work follows a generic workflow for compression which I believe can be useful for future rapid development and testing.

The below workflow allows for a base accumulator or multiple accumulators, multiple predictors using different accumulators (models). Those predictors would be put through a mixer or number of mixers. After the mixers produce a final refined prediction of the best guess to the encoder or decoder (depending on the input being a compressed or uncompressed file) and outputs the code or original information and continues along the process back to the accumulator.

What are the benefits of this type of model ? So far it allows for rapid testing of different mixing strategies, different models for predicting, and different rates of change for accumulating statistics. This workflow also allows for the creation of a neural network mixed with traditional compression techniques.

An example of this beneficial strategy is that adding a predictor has no impact or change on the accumulator, multiple predictors can be swapped out or tested at anytime. Changing the accumulator will however have an impact on the predictor using that accumulator as well as the mixer.

X-Hex is moving along. I have updated the graphics and started production on my new IMAC. Unfortunately there is a bug with the production software that is causing a delay. I would like to get everything ready to post another video soon (It’s been two years). Let me know what you think.

A little preview of the Grid used in X-Hex. With over 40 3D universes, dynamic HD lighting, and physically based rendering, X-hex has an amazing visual and varied atmosphere.

Click on the images below for a full HD preview.

Backdrop

Backdrop

Backdrop

I wrote this piece of software to compliment my article about IT Metrics (https://chrischunick.com/?p=142). This software allows the process of gathering statistics to be enhanced through the efficiency of automation to quickly submit a ticket without having to login to a ticketing system. The program interfaces with a backend (currently spiceworks), and allows the end user to submit multiple tickets one after the other or simply shrink the software to be left out of sight until another ticket comes up.  The current user is the logged in user, email is parsed from Active Directory.

Tickets can be written down throughout the day and entered at a later time, entered as an email,  entered as a phone call comes in, or if an individual walks in requesting service. One of the challenges in IT is to maintain a consistent workflow while also being able to document the work without interruptions. This software has so far been tested successfully in a production environment.