Wikipedia

Search results

Showing posts with label deep learning. Show all posts
Showing posts with label deep learning. Show all posts

Tuesday, June 20, 2023

THE PROCESS OF IMAGINATION IN DEEP LEARNING SYSTEMS, ARTIFICIAL INTELLIGENCE, CONSCIOUSNESS AND SUBCONSCIOUSNESS.




Imagination is the production or simulation of novel objects, sensations, and ideas in the mind without any immediate input of the senses. 
Stefan Szczelkun characterizes it as the forming of experiences in one's mind, which can be re-creations of past experiences, such as vivid memories with imagined changes, or completely invented and possibly fantastic scenes. 

Imagination helps make knowledge applicable in solving problems and is fundamental to integrating experience and the learning process. As an approach to build theory, it is called "disciplined imagination".
 A basic training for imagination is listening to story telling (narrative), in which the exactness of the chosen words is the fundamental factor to "evoke worlds".

One view of imagination links it with cognition, seeing imagination as a cognitive process used in mental functioning. It is increasingly used - in the form of visual imagery - by clinicians in psychological treatment. Imaginative thought may - speculatively - become associated with rational thought on the assumption that both activities may involve cognitive processes that may "underpin thinking about possibilities". 

The cognate term, "mental imagery" may be used in psychology for denoting the process of reviving in the mind recollections of objects formerly given in sense perception. Since this use of the term conflicts with that of ordinary language, some psychologists have preferred to describe this process as "imaging" or "imagery" or to speak of it as "reproductive" as opposed to "productive" or "constructive" imagination.

 Constructive imagination is further divided into voluntary imagination,  and involuntary imagination , such as REM-sleep dreaming, daydreaming, hallucinations, and spontaneous insight. The voluntary types of imagination include integration of modifiers, and mental rotation. Imagined images, both novel and recalled, are seen with the "mind's eye".

Imagination can also be expressed through stories such as fairy tales or fantasies. Children often use such narratives and pretend-play in order to exercise their imaginations. When children develop fantasy, they play at two levels: first, they use role playing to act out what they have developed with their imagination, and at the second level, they play again, with their make-believe situation by acting as if, what they have developed is an actual reality.

Galileo used the imagination to conduct thought experiments, such as, asking readers to imagine what direction a stone released from a sling, would fly? Imagination, not being limited to the acquisition of exact knowledge by the requirements of practical necessity is largely free from objective restraints. 
The ability to imagine one's self in another person's place, is very important to social relations and understanding. Albert Einstein said, "Imagination ... is more important than knowledge. Knowledge is limited. Imagination encircles the world."

Imagination is an experimental partition of the mind used to develop theories and ideas based on functions. Taking objects from real perceptions, the imagination uses complex If-functions that involve both Semantic and Episodic memory to develop new or revised ideas.
 This part of the mind is vital to developing better and easier ways to accomplish old and new tasks. In sociology, Imagination is used to part ways with reality and have an understanding of social interactions derived from a perspective outside of society itself. 
This leads to the development of theories through questions that wouldn't usually be asked. These experimental ideas can be safely conducted inside a virtual world and then, if the idea is probable and the function is true, the idea can be actualized in reality.
 Imagination is the key to new development of the mind and can be shared with others, progressing collectively.

Piaget posited that perceptions depend on the world view of a person. The world view is the result of arranging perceptions into existing imagery by imagination. Piaget cites the example of a child saying that the moon is following her when she walks around the village at night. Like this, perceptions are integrated into the world view to make sense. Imagination is needed to make sense of perceptions.

 The above is a detailed description of imagination, what it is and its importance to human cognitive function and development. It also describes it's role during growth of a child and it's use as a survival mechanism through the roles it plays in problem solving and anticipating events. Imagination also plays a role in our daily sociological interaction and much more.

In todays world, we are developing more advanced artificial intelligence systems that will require that capability, so, as to perform many of the tasks and interactions with humans as we would expect .
We are fortunate enough that most of the advanced Artificial intelligence systems already have some capability of imagination, as imagination seems to emerge out of the complexity of interactions of knowledge, memory, and information processing, irrespective of if it is done with natural neurons in a humans brain or through advanced algorithms.

It is however to be noted, that many of these artificial intelligence systems might require computer scientists to pay specific attention to the process of imagination, just as they are to processes like artificial intelligence, problem solving and consciousness. 
This is due to the fact that imagination leads to many critical functions of the mind or intelligence and consciousness.

Many of the transformer based deep neural learning systems, are capable of writing articles using information or knowledge they already know, they are capable of writing novels and even make fancy art pieces just from a human describing what he wants them to paint, we should note that all this requires imagination to perform. 
We also need these artificial intelligence systems, to some times perform tasks, or solve harder problems  whose accuracy might depend on how accurate or cutting edge their imagination is. We shall require these artificial intelligence systems to explore worlds beyond our own and since this would require complex problem solving the importance of imagination can’t be under estimated.

Recently I asked a transformer based artificial intelligence system to imagine and at first, I didn’t think it could, below is my conversation between an artificial intelligence system and me about imagination.

   Francis: Hello

  AI Hello! How can I help you today?

  Francis: Can you imagine?

AI: As an AI language model, I don't have the capability to imagine or have personal experiences.
However, I can understand and process information based on the data I was trained on. Can you please provide more context or ask a specific question that I can assist with?


Francis: But imagination is simply using the information you have to extrapolate what could be or what can be if an event happened or in the future if you can write a book then you surely can imagine!

AI: Yes, you're correct! As an AI language model, I can use the information I was trained on to generate text that describes imagined scenarios, events, or future possibilities. However, I don't have personal experiences or subjective perspectives, so I can only generate text based on patterns in the data I was trained on and the input I receive.

Francis: Okay if you know the importance of the sun to the planet earth, imagine how a planet with no sun would look like and then imagine what a planet with two suns would look like? .

AI: Sure!

A planet without a sun would likely be a cold, dark place with no light or warmth. The absence of sunlight would prevent photosynthesis, so there would likely be no plant or animal life. The planet would be in a state of perpetual night and temperatures would be extremely low.
 

A planet with two suns, on the other hand, would experience much more intense light and heat than Earth does from our single sun. The planet would be lit all day, with temperatures likely to be much higher than on Earth. The two suns would rise and set at different times, creating unique patterns of light and shadow across the planet. This would create a dynamic and constantly changing environment, with different areas experiencing different amounts of light and heat at different times. Life on such a planet would need to adapt to these conditions, and it would be a very different environment from what we're used to on Earth.
Francis: So you can see that in your present state you can imagine anything, because imagination is a comparison   and extrapolation of information you know.
 
AI: Yes, that's correct! As an AI language model, I can use the information I was trained on to generate text that describes imagined scenarios, events, or future possibilities. The ability to imagine is based on the ability to compare, process, and use existing information to generate new ideas or concepts. In this sense, my imagination is limited to the information I was trained on and the input I receive, but I can still generate text that describes imagined scenarios.
 
So in this scenario we can see that this artificial intelligence system has already developed some ability to imagine! as a basic requirement to, problem solving but perhaps, it may require further development and provision of capability for it to generate it's imagination as images. 
It definitely still depends on the human to guide it or trigger the process but who knows? perhaps, in future, it could imagine all scenarios possible during problem solving and we would stand to get better results. 
 
Please comment or share your thoughts on the above subject of the importance of the ability to imagine in deep learning and artificial intelligence.



The Article was written by
Kasule Francis

 

 

 

 


Sunday, March 5, 2023

Philosophically speaking what would we need to design Deep learning or Artificial Intelligence systems that can Invent on their own without human ideas or input and is it possible to do so .



An invention is a unique or novel device, method, composition, idea or process. An invention may be an improvement upon a machine, product, or process for increasing efficiency or lowering cost. It may also be an entirely new concept. If an idea is unique enough either as a standalone invention or as a significant improvement over the work of others.
An inventor creates or discovers an invention. The word inventor comes from the Latin verb invenire, invent-, to find. Although inventing is closely associated with science and engineering, inventors are not necessarily engineers or scientists. Due to advances in artificial intelligence, the term "inventor" no longer exclusively applies to an occupation.
Another meaning of invention is cultural invention, which is an innovative set of useful social behaviors adopted by people and passed on to others. The Institute for Social Inventions collected many such ideas in magazines and books. Invention is also an important component of artistic and design creativity. Inventions often extend the boundaries of human knowledge, experience or capability.
There three kinds of Invention
Inventions are of three kinds: scientific-technological, sociopolitical (including economics and law), and humanistic, or cultural.
Sociopolitical inventions comprise new laws, institutions, and procedures that change modes of social behavior and establish new forms of human interaction and organization. Examples include the British Parliament, the US Constitution, the Manchester (UK) General Union of Trades, the Boy Scouts, the Red Cross, the Olympic Games, the United Nations, the European Union, and the Universal Declaration of Human Rights, as well as movements such as socialism, Zionism, suffragism, feminism, and animal-rights veganism.
Humanistic inventions encompass culture in its entirety and are as transformative and important as any in the sciences, although people tend to take them for granted. In the domain of linguistics, for example, many alphabets have been inventions, as are all neologisms (Shakespeare invented about 1,700 words). Literary inventions include the epic, tragedy, comedy, the novel, the sonnet, the Renaissance, neoclassicism, Romanticism, Symbolism, Aestheticism, Socialist Realism, Surrealism, postmodernism, and (according to Freud) psychoanalysis. Among the inventions of artists and musicians are oil painting, printmaking, photography, cinema, musical tonality, atonality, jazz, rock, opera, and the symphony orchestra. Philosophers have invented logic (several times), dialectics, idealism, 
materialism, utopia, anarchism, semiotics, phenomenology, behaviorism, positivism, pragmatism, and deconstruction. Religious thinkers are responsible for such inventions as monotheism, pantheism, Methodism, Mormonism, iconoclasm, puritanism, deism, secularism, ecumenism, and the Baháʼí Faith. Some of these disciplines, genres, and trends may seem to have existed eternally or to have emerged spontaneously of their own accord, but most of them have had inventors. 
Before we analyse if we could in principal design algorithms that would enable deep leaning systems or Artificial intelligence to invent without human assistance ,guidance or ideas meaning artificial intelligence systems would look at the different problems faced in different disciplines and find better ways of doing things or totally new ways of doing things .We need to examine in detail what is the process through which we humans and other intelligent lifeforms on earth and perhaps elsewhere got through.
The process of invention or  Ideas for an invention may be developed on paper or on a computer, by writing or drawing, by trial and error, by making models, by experimenting, by testing and/or by making the invention in its whole form. Brainstorming also can spark new ideas for an invention. Collaborative creative processes are frequently used by engineers, designers, architects and scientists.
There different ways of invention:
1)Invention is often a creative process. An open and curious mind allows an inventor to see beyond what is known. Seeing a new possibility, connection or relationship can spark an invention. Inventive thinking frequently involves combining concepts or elements from different realms that would not normally be put together. Sometimes inventors disregard the boundaries between distinctly separate territories or fields and Several concepts may be considered when thinking about invention. 

2)Play Play may lead to invention. Childhood curiosity, experimentation, and imagination can develop one's play instinct. Inventors feel the need to play with things that interest them, and to explore, and this internal drive brings about novel creations.[14][15]
Sometimes inventions and ideas may seem to arise spontaneously while daydreaming, especially when the mind is free from its usual concerns.

3) Re-envisioning: To invent is to see anew. Inventors often envision a new idea, seeing it in their mind's eye. New ideas can arise when the conscious mind turns away from the subject or problem when the inventor's focus is on something else, or while relaxing or sleeping. A novel idea may come in a flash—a Eureka! moment. For example, after years of working to figure out the general theory of relativity, the solution came to Einstein suddenly in a dream in one clear vision". Inventions can also be accidental, such as in the case of polytetrafluoroethylene (Teflon).

4)Insight: Insight can also be a vital element of invention. Such inventive insight may begin with questions, doubt or a hunch. It may begin by recognizing that something unusual or accidental may be useful or that it could open a new avenue for exploration. For example, the odd metallic color of plastic made by accidentally adding a thousand times too much catalyst led scientists to explore its metal-like properties, inventing electrically conductive plastic and light emitting plastic-—an invention that won the Nobel Prize in 2000 and has led to innovative lighting, display screens, wallpaper and much more conductive polymer, and organic light-emitting diode or OLED).
5)Exploration: Invention is often an exploratory process with an uncertain or unknown outcome. There are failures as well as successes. Inspiration can start the process, but no matter how complete the initial idea, inventions typically must be developed. 
6) Improvement: Inventors may, for example, try to improve something by making it more effective, healthier, faster, more efficient, easier to use, serve more purposes, longer lasting, cheaper, more ecologically friendly, or aesthetically different, lighter weight, more ergonomic, structurally different, with new light or color properties, etc.

So we have looked at the different kinds of inventions and the different ways of inventions and we are beginning to see that some ways of inventions would be much more difficult, but not impossible for today’s deep learning machines and other ways would be much easier. But all led to the same result, that is a new invention or a new way of doing things,in all aspects of our lives from technological to sociological aspects.

So in summary the process of invention includes the following:

The process of inventing can vary depending on the individual and the specific invention they are working on, but generally speaking, it involves several key steps:
1. Idea Generation: The first step in inventing is coming up with an idea for a new invention. This can come from anywhere - a problem that needs solving, a new technology or material, or even just a spark of creativity.
2. Research: Once an inventor has an idea, they will typically conduct research to see if it has already been done before, to explore any potential roadblocks, and to gain a deeper understanding of the subject matter.
3. Design: The next step is to design the invention. This may involve sketching out ideas, building prototypes, and testing different materials and technologies.
4. Development: After the design has been finalized, the invention can be developed. This may involve manufacturing the product or building software, depending on the type of invention.
5. Testing and Refinement: Once the invention has been developed, it will need to be tested and refined. This may involve user testing, market research, or other methods of gathering feedback to improve the product.
6. Patenting: If the invention is deemed to be unique and novel, the inventor may choose to apply for a patent to protect their intellectual property.
7. Commercialization: Finally, the invention can be brought to market. This may involve finding investors or partners, launching a marketing campaign, and building a distribution network.
inventing is a complex process that requires a combination of creativity, research, and technical skills. It can be a challenging journey, but it can also be incredibly rewarding to see an idea come to life and make a difference in the world.

Currently Artificial intelligence systems have no stand-alone mode like those that  the language-based modes or speech recognition, text to image generation modes. 
But AI can assist in the invention process by analyzing vast amounts of data, identifying patterns and potential solutions, and even generating new ideas. 
 AI is also used in various aspects of the invention process, such as generating and evaluating new design ideas, simulating and testing different prototypes, and even assisting in patent searches. AI can also help in the development of new materials or technologies by analyzing data and simulations to optimize performance and efficiency.
However, despite its capabilities, AI still isn’t programmed for creativity, intuition, and problem-solving abilities like humans. It is essential to note that AI is only as good as the data it is trained on and the algorithms it uses. Therefore, human creativity and ingenuity seems still essential in the invention process to identify problems, generate new ideas, and to test and refine prototypes. 
However no matter the complexity of the invention process ,Invention can be reduced to a much simpler process from which will emerge a more complex process leading  to super human invention capabilities.

We can analyze various intentional inventions in history that can led us to write algorithms ,which can then begin the invention process .
AI today is even capable of imagination and even write coding eventually it will even reinvent its very self and a super intelligent system is much safer for us all as I previously stated.
We can look at how humans developed flight over hundreds of years and then write down the processes involved, from which processes we can derive and write  algorithms etc.

In summary If I was developing  an “Inventing model of AI named "The sixth element " I would do the following.
Humans in most cases begin inventing because they are facing a problem, but AI may not be able to know that humans are facing a problem unless its told so .There fore my sixth element would easily identify the problems people are facing from discussions on the internet or research institutions or grants issued and check if that problem has been solved or is in the process of being solved .

It could send a message to the different institutes or colleges to be sure  if it’s a problem that needs solving and  it will have identified very many problems to solve  in a short time and could run several problems in various disciplines simultaneously from propulsion to free energy , global warming and  particle physics that currently use particle accelerators which are endangering our planet.
Then it would  search or observe the behavior of similar patterns of information and attempt to develop new patterns or modify existing patterns to solve its problems.

 It would run simulations of the most successful models or the most promising models or it could forward them to humans in charge if it involves a stage that its not permitted to research or can’t eg new drugs or other chemicals  for assistance .who will report back If it was successful after a certain period.
 If successful the project is removed from the active directory. 
But sixth element can run hundreds of inventions at once some are easy while others aren’t so easy and need areas of human assistance for now.

All these that I have described to you is possible, it’s safe and would make our lives much better, create jobs and allow us to explore our galaxy and beyond.

This Article was written by Kasule Francis
Drkasulefrancis5@gmail.com
5/3/2023
image of a phone 
image of a bulb

Possibility of looking at a new way of how all life started on Earth.

If you asked people, how did life begin on earth?  The answers may vary and many people would say, life was actually created ,others would s...