Throughout my career both as a physicist and as a quant in the business world I have built simulators. Like most other simulation experts I first develop a mathematical model and then implement the model as a computer program. The term “mathematical model” includes many different kinds of models. Continuous functional forms expressing physical laws are examples of mathematical models. Statistical models (parametric as well as non-parametric), algorithms and all kinds of machine learning techniques also qualify as “mathematical model.”
Layers of a simulator
Excluding the input/output layers a simulator consists of 3 layers
- The physical substrate. In the case of simulators implemented as computer programs the physical substrate is the electronic circuitry of the computer.
- The computer program which exists as a an organized set of 1 and 0’s in the memory section of the computer.
- The “mathematical model” encoded in the computer program. The “mathematical model” is hidden in the organized set of 1 and 0’s.
A very important point to note here is that the third layer – the mathematical model – does not emerge from the substrate and the computer program. The third layer is put in by an intelligent agent. Someone has to implement the mathematical model in the computer program and place the program in the physical substrate – the electronic circuitry. The third layer of a simulator is not an emergent phenomenon. Remember this.
Dreaming about a natural simulator
I have always dreamed about finding a way to build a natural simulator that does not use the digital paradigm. I have used the digital (1 and 0’s) programming to the best of my abilities and witnessed the limitations of this paradigm first hand.
Where do I get this idea of a “natural simulator?” I have been studying Eastern teachings for more than 30 years. One of the fascinating concepts taught in the East is the “mind stuff” (“citta” in Sanskrit).
Citta is the densest part of the mind. It has its own nature. We don’t know the laws of citta but we are aware of its first and foremost characteristic: citta-brain combination is the ultimate simulator.
Citta is the third layer of this ultimate simulator. In other words, citta is similar to the “mathematical model” of a typical simulator implemented in the digital programming paradigm. The so called “citta form” is similar to a mathematical model.
Since we don’t know much about citta can we simulate its known behavior to gain more understanding? In other words, can we simulate “higher mathematical” model with a “lower mathematical” model? The term “higher mathematical” is my way of imagining the “citta form.”
We are not talking about simulating the brain function. Brain corresponds to the first layer of the citta-brain complex. We are not talking about simulating the second layer either (hormones, proteins, DNA, millions of chemical mechanisms as well as the vital energy flows = 10 vayus – that are not yet known to Western science).
We are taking about simulating the third layer (citta medium and citta-forms). As I mentioned earlier the citta-forms can be described as “higher mathematical” models. I am sure that we will expand the definition of “mathematical” in the future to cover the modalities of citta-forms. We did something similar recently when we included algorithms and machine learning techniques in the definition of “mathematical.”
Can we simulate “higher mathematical” model with a “lower mathematical” model? My intuition tells me that it is possible. I am sorry I don’t have a concrete example here but I know intuitively that we can someday simulate the citta. That would also be the beginning of the era of natural simulators.
The “natural simulator” would be different from analogue computers and quantum computers.
In electrical engineering the term “analogue” refers to devices that work by taking advantage of specific physical laws. Analogue devices are special designs manipulating nature. The “digital computing paradigm” on the other hand is independent of nature in the sense that we can employ different physical laws to implement the “bit.” The bit represents the “yes/no” (1 or 0). The “digital computing paradigm” is built on the “bit.” The “analogue” device, on the other hand, is totally dependent on the specific functioning of the physical laws it is employing. An analogue computer would be unique by definition.
The quantum computing paradigm is a marriage of the “analogue” and the “digital.” The quantum computing is “analogue” in the sense that it takes advantage of a specific physical law – quantum physics. The quantum computing is also “digital” in the sense that it has the concept of a “qubit” similar to the concept of “bit” of digital computing.
Information atom more powerful than bit or qubit
The “natural simulator” would not be based on specific natural laws. This means that the “natural simulator” would not be an “analogue” computer. This rules out the similarity to quantum computing as well because quantum computers use a specific natural law – quantum mechanics. The “natural simulator” would not be a “digital” computer either.
The “natural simulator” would NOT be based on “bit” or “qubit” but another kind of information atom that has much more capacity to represent information. If you are familiar with my writing you could guess my candidate for the information atom. I will write about information atoms in the future.