Artificial Intelligence And Intuition

 

The intuitive algorithm

Roger Penrose considered it impossible. Thinking could never imitate a computer process. He said as much in his book, The Emperor's New Mind. But a new book, The Intuitive Algorithm (IA), suggested that intuition was a pattern recognition process. Intuition propelled information through many neural regions like a lightning streak. Data moved from input to output in a reported 20 milliseconds. The mind saw, recognized, interpreted, and acted. In the blink of an eye. Myriad processes converted light, sound, touch, and smell instantly into your nerve impulses. A dedicated region recognized those impulses as objects and events. The limbic system, another region, interpreted those events to generate emotions. A fourth region responded to those emotions with actions. The mind perceived, identified, evaluated, and acted. Intuition got you off the hot stove in a fraction of a second. It could operate using a simple algorithm.

Is instant holistic evaluation impossible?

The system, with over a hundred billion neurons, processed the information from input to output in just half a second. All your knowledge was evaluated. Walter Freeman, a renowned neurobiologist, described this remarkable ability. "The cognitive experts believe it's simply unfeasible to consistently contribute all of your resources to the computation." But that is exactly what the brain does. Consciousness is about bringing your entire history to bear on your next step, your next breath, and your next moment." The mind was holistic. It evaluated all its knowledge for the next activity. How was it possible to process so much information so quickly? Where could such knowledge be stored? 

Exponential growth in the search path 

Unfortunately, the recognition of subtle patterns posed formidable problems for computers. The challenge arose from the exponential growth of the search path for recognition. The challenges encountered during disease diagnosis were typical. Normally, many shared symptoms were presented by a multitude of diseases. For instance, various diseases could indicate pain or fever. Each symptom pointed to several diseases. The challenge was to identify a single pattern among numerous overlapping patterns. The first disease chosen may not have the second symptom when searching for the target disease. This meant back and forth searches, which expanded exponentially as the database of diseases increased in size. This resulted in an absurdly drawn-out process, potentially taking years to search through extensive databases. Hence, one could never imagine rapid pattern recognition on computers, despite their incredible speed.

The intuitive algorithm

However, industrial strength pattern recognition was feasible. IA introduced an algorithm that could instantly recognize patterns in extended databases. The relationship of each member of the whole database was coded for each question.

Is pain a symptom of the disease?

Disease 1Y, Disease 2N, Disease 3Y, Disease 4Y, Disease 5N, Disease 6N, Disease 7Y, Disease 8N, Disease 9N, Disease 10N, Disease 11Y, Disease 12Y, Disease 13N, Disease 14U, Disease 15Y, Disease 16N, Disease 17Y, Disease 18N, Disease 19N, Disease 20N, Disease 21N, Disease 22Y, Disease23N, Disease24N, Disease 25U, Disease26N, Disease 27N, Disease28U, Disease27Y, Disease 30N, Disease31U, Disease32Y, Disease33Y, Disease 34U, Disease 35N, Disease 36U, Disease 37Y, Disease 38Y, Disease 39U, Disease 40Y, Disease 41Y, Disease 42U, Disease 43N, Disease 44U, Disease45Y, Disease46N, Disease47N, Disease48Y, 

(Y = Yes; N = No; U = Uncertain)

The key was to use elimination to evaluate the database, not selection. We individually coded each database member for elimination in the context of each answer.

Is pain a symptom of the disease? Answer: YES)

Disease1Y, xxxxxxN, Disease3Y, Disease4Y, xxxxxx5N, xxxxxx6N, Disease7Y, xxxxxx8N, xxxxxx9N, xxxxxx0N, Disease11Y, Disease12Y, xxxxxx13N, Disease14U, Disease15Y, xxxxxx16N, Disease17Y, xxxxxx18N, xxxxxx19N, xxxxxx20N, xxxxxx21N, Disease22Y, xxxxxx23N, xxxxxx24N, Disease25U, xxxxxx26N, xxxxxx27N, Disease28U, xxxxxx30N, Disease31U, Disease32Y, Disease33Y, Disease34U, xxxxxx35N, Disease36U, Disease37Y, Disease38Y, Disease39U, Disease40Y, Disease41Y, Disease42U, xxxxxx43N, Disease 44U, Disease45Y, xxxxxx46N, xxxxxx47N, Disease 48Y

(All "N" diseases eliminated.)

For disease recognition, if an answer indicated a symptom, IA eliminated all diseases devoid of the symptom. We eliminated every answer, thereby narrowing the search to a diagnosis.

Is pain a symptom of the disease? Answer: NO)

xxxxxx1Y, Disease2N, xxxxxx3Y, xxxxxx4Y, Disease5N, Disease6N, xxxxxx7Y, Disease8N, Disease9N, Disease10N, xxxxxx11Y, xxxxx12Y, Disease13N, Disease14U, xxxxxx15Y, Disease16N, xxxxxx17Y, Disease18N, Disease19N, Disease20N, Disease21N, xxxxxx22Y, Disease23N, Disease24N, Disease25U, Disease26N, Disease27N, Disease28U, xxxxxx27Y, Disease30N, Disease31U, xxxxxx32Y, xxxxxx33Y, Disease34U, Disease35N, Disease36U, xxxxxx37Y, xxxxxx38Y, Disease39U, xxxxxx40Y, xxxxxx41Y, Disease42U, Disease43N, Disease 44U, xxxxxx45Y, Disease46N, Disease47N, xxxxxx48Y, 

(All "Y" diseases eliminated.)

If the symptom was absent, IA eliminated all diseases that always exhibited it. Both cases retained diseases that presented the symptom randomly. So the process handled uncertainty—the “Maybe” answer, which normal computer programs could not handle.

The sequence of questions narrows down to Disease 29—the answer.

xxxxxx1Y, xxxxxx2N, xxxxxx3Y, xxxxxx4Y, xxxxxx5N, xxxxxx6N, xxxxxx7Y, xxxxxx8N, xxxxxx9N, xxxxxx10N, xxxxxx11Y, xxxxxx12Y, xxxxxx13N, xxxxxx14U, xxxxxx15Y, xxxxxx16N, xxxxxx17Y,xxxxxx18N, xxxxxx19N, xxxxxx20N, xxxxxx21N, xxxxxx22Y, xxxxxx23N, xxxxxx24N, xxxxxx25U, xxxxxx26N, xxxxxx27N, xxxxxx28U, Disease29Y, xxxxxx30N, xxxxxx31U, xxxxxx32Y, xxxxxx33Y, xxxxxx34U, xxxxxx35N, xxxxxx36U, xxxxxx37Y, xxxxxx38Y, xxxxxx39U, xxxxxx40Y, xxxxxx41Y, xxxxxx42U, xxxxxx43N, xxxxxx44U, xxxxxx45Y, xxxxxx46N, xxxxxx47N, xxxxxx48Y. 

When we eliminate all diseases, the disease remains unknown.

Instant pattern recognition 

IA was proved in practice. Expert Systems acted with the speed of a simple recalculation on a spreadsheet to recognize a disease, identify a case law, or diagnose the problems of a complex machine. It was instant, holistic, and logical. Recognition was instantaneous when presented with multiple parallel answers, such as the multiple parameters of a power plant. Real-time pattern recognition was practical for the mind, which presented millions of parameters simultaneously. And elimination was the key.

Elimination means switching off.

Elimination was switching off—inhibition. It was known that nerve cells extensively inhibit the activities of other cells to highlight context. With access to millions of sensory inputs, the nervous system instantly inhibited and eliminated trillions of combinations to pinpoint the correct pattern. The process stubbornly used "no" answers. A doctor could ignore thousands of possible diseases if a patient did not experience pain. If a patient could just walk into the surgery, a doctor could overlook a wide range of illnesses. How can we apply this process of elimination to nerve cells? Where could we store the wealth of knowledge?

Combinatorial coding

The mind received kaleidoscopic combinations of millions of sensations. Researchers reported that nerve cells recognized smells through a combinatorial coding process. If a nerve cell had dendritic inputs, identified as A, B, C, and so on to Z, it could then fire when it received inputs at ABC, or DEF. It recognized those combinations. The cell could identify ABC and not ABD. It would be inhibited for ABD. Science recently reported this recognition process for olfactory neurons. In the experiment, scientists reported that even slight changes in chemical structure activated different combinations of receptors. Thus, octanol smelled like oranges, but the similar compound octanoic acid smelled like sweat. A Nobel Prize acknowledged that discovery in 2004. 

Galactic nerve cell memories

Nature extensively used combinatorial codes. Nature used combinations of the four "letters" in the genetic code—A, C, G, and T—to create a nearly infinite number of genetic sequences. IA discusses the deeper implications of this coding discovery. Animals could differentiate between millions of smells. Dogs could quickly sniff a few footprints of a person and determine accurately which way the person was walking. The animal's nose could detect the relative odor strength difference between footprints only a few feet apart to determine the direction of a trail. Smell was identified through remembered combinations. If a nerve cell had just 26 inputs from A to Z, it could receive millions of possible combinations of inputs. The average neuron had thousands of inputs. For IA, millions of nerve cells could provide the mind with galactic memories of combinations, enabling it to recognize subtle patterns in the environment. Each cell could be a single member of a database, eliminating itself (becoming inhibited) for unrecognized combinations of inputs.

Elimination is key.

The key to evaluating vast combinatorial memories was elimination. Medical texts reported that the mind had a hierarchy of intelligences, which performed dedicated tasks. For example, there was an association region that recognized a pair of scissors using the context of its feel. If you injured this region, you would still be able to feel the scissors with your eyes closed, but your ability to recognize them as scissors would be impaired. You still felt the context, but you would not recognize the object. So, intuition could enable nerve cells in association regions to use perception to recognize objects. Medical research reported many such recognition regions. 

Serial processing

A pattern recognition algorithm and intuition enabled the finite intelligences in the minds of living things to respond holistically within the 20-millisecond time span. These intelligences acted serially. The first intelligence converted the kaleidoscopic combinations of sensory perceptions from the environment into nerve impulses. The second intelligence recognized these impulses as objects and events. The third intelligence translated the recognized events into feelings. A fourth translated feelings into intelligent drives. Fear triggered an escape drive. A deer bound away. A bird took flight. A fish swam off. While the activities of running, flying, and swimming differed, they achieved the same objective of escaping. Inherited nerve cell memories powered those drives in context.

The mind: seamless pattern recognition

It takes half a second for 100 billion nerve cells to utilize context, remove irrelevant information, and produce motor output. The interval between a shadow's appearance and a scream's utterance is significant. So, from input to output, the mind was a seamless pattern recognition machine, powered by the key secret of intuition—contextual elimination—from massively acquired and inherited combinatorial memories in nerve cells.

Location:

Share :