Global Education Platform for Global Literacy: Global AI Academy: AI4EE: AI Technology for Universal Education to the New Man-Machine Economy
Ensure inclusive and equitable quality education and promote lifelong learning opportunities for all (SDG 4 )
Being illiterate about the basics of artificial intelligence in the 21st century is the same being illiterate with reading, writing and arithmetic.
Introduction: What is Global Literacy or Illiteracy?
As to the Office of the UN Special Envoy for Global Education , our world is afflicted by a global skills gap. The need for skilled individuals is growing rapidly while the young and marginalised lack access to the education that would allow them to meet this need.?
The global literacy rate currently stands at?87%, up from 12% in 1820 when the world population was about 1bn.? There are now 773 million?illiterate adults around the world, most of whom are women.
Today, there are only around?220 million?tertiary education students in the world , ?as referring to all formal post-secondary education, including public and private universities, colleges, technical training institutes, and vocational schools. The economic returns for tertiary education graduates have only an estimated 17% increase in earnings as compared with 10 % for primary and 7% for secondary education; for many do not have relevant knowledge and skills needed for a successful integration into the new challenging labor market of the 21st century.
The entire educational system – from early childhood through tertiary education – must reflect the new social and economic needs of the global knowledge economy, which increasingly demands a more knowledgeable, better-trained, more skilled, adaptable and AI-literate workforce.?
There must be the comprehensive schools for all ISCED 2011 levels of education and above:
primary school, elementary education (fundamental reading, writing, and mathematics skills and establish a solid foundation for learning)
secondary education (academy schools, community schools, faith schools, foundation schools, free schools, studio schools, university technical colleges, state boarding schools, City Technology Colleges, etc.)
tertiary education (third-level,?third-stage?or?post-secondary education)
life-learning education , as all learning activity undertaken throughout life, with the aim of improving knowledge, skills and competences within a personal, civic, social and/or employment-related perspective", for all (SDG4).
What is missing UNESCO and national educational institutions?
The main objective of UNESCO is "to contribute to peace and security in the world by promoting collaboration among nations through education, science, culture and communication in order to foster universal respect for justice, the rule of law, and the human rights and fundamental freedoms that are affirmed for the peoples of the world, without distinction of race, sex, language or religion, by the Charter of the United Nations".?
National education systems vary in terms of structure and curricular content, it can be difficult to benchmark performance across countries over time or monitor progress towards national and international goals.
The International Standard Classification of Education (ISCED) belongs to the United Nations International Family of Economic and Social Classifications, which are applied in statistics worldwide with the purpose of assembling, compiling and analysing cross-nationally comparable data. ISCED is the reference classification for organizing education programmes and related qualifications by education levels and fields.??
What is missing in all the modern education is a systematic and systemic teaching of global literacy basing on multi-, inter- and trans-disciplinary education, science and technology.
What are the elements of global literacy?
Types of literacy/illiteracy:
Global Emerging Technology AI Illiteracy
The criterion of modern education, knowledge and skills is emerging technologies led by artificial intelligence, its science, engineering and technology. As to WEF, Without universal AI literacy, AI will fail us .
AI is everywhere — whether we’re aware of it or not.
From displacement and hunger to infectious disease outbreaks and climate change, this technology has the potential to help us tackle some of the toughest global challenges. In fact, AI could enable the accomplishment of?134 targets ?— out of 169 — across all U.N. Sustainable Development Goals. Besides,
A lack of transparency, awareness, and understanding of AI among the general population is merely intolerable, a?national survey ?found that 84% of Americans are illiterate about AI.
Many national universities employ such universal AI illiteracy for their commercial benefits, creating low-quality AI graduate programs, as extension of their CS departments or Engineering Depts or Statistics Depts or Data Science Depts or Math Depts. Here is the common case of Imperial College London with the core MSc core modules:
There are a number of caveats which concerns the validity of most universities AI programs.
AI?is wrongly studied as “the simulation of human intelligence processes by machines, especially computer systems, with specific applications as symbolic AI, expert systems, natural language processing, speech recognition or machine vision”.
As a programmer, you will be taught to teach/train/program such a human-like AI/ML/DL to automatically identify correlation patterns in datasets that may contain structured data, quantitative data, textual data, visual data, etc.
Just keep in the mind the real prospects of Real AI not as a Statistical Deep Learning AI, but as a Causal Learning AI Machine (CLAIM).
The CLAIM Top Functions are to intelligently identify causal patterns in the data universe of various data sets.
Its top requirements are much more challenging than the university AI:
Reality/Causality/Science > World Learning and Inference Model > Statistics/Data Science > Mathematics/Set Theory/Optimization/Calculus/Linear Algebra/Probability > Programming/Algorithms/Neural Networks > Causal Regression > Real-World Applications
Caveat Emptor, Let the buyer beware.
Global AI Academy (GAIA) Education Platform: AI4EE
The GAIA is to advance the highly-innovative socio-technological educational agenda:
The COVID-19 crisis?has accelerated the need for new knowledge, competences and workforce skills , advanced cognitive, scientific, technological and engineering, and social and emotional skills.
In the AI and Robotics era, there is a high demand to the scientific knowledge, digital competence and high-technology training in a range of innovative areas of exponential technologies, such as artificial intelligence, machine learning and robotics, data science and big data, cloud and edge computing, the Internet of Thing, 6G, cybersecurity and digital reality.
The European Round Table for Industry launched a pan-European training initiative to help unemployed and at-risk workers. The Reskilling 4 Employment aims to reskill one million workers by 2025, and up to five million by 2030. Initial pilot projects are planned in Portugal, Spain, and Sweden, and corporate supporters include AstraZeneca, Iberdrola, Nestlé, SAP, Sonae, and Volvo Group.
The European Network of AI Excellence Centres are to drive up collaboration in research across Europe, bringing together world-class researchers and establishing a common approach, vision and identity for the European AI ecosystem. The initiative aims the following:
Another example is Cisco’s Networking Academy. The company partners with educators and instructors around the world to offer students IT training in a range of areas such as big data, cloud, cybersecurity, and machine learning. The effort connects students to jobs inside Cisco and with its external partners, while creating a much larger pool of skills the company prioritizes. Besides, Cisco Systems teamed up with universities to create centralized data science and AI training programs units for employees to transform them into experts.
Most AI & ML Academies and CoEs are restrained by an anthropocentric Narrow and Weak AL & ML, referring it "to a?suite of technologies and methods that seek to simulate characteristics associated with intelligence in humans or other living systems. Machine learning is an approach to AI in which models process data, learning from this data to identify patterns or make predictions". This affects the whole enterprise in the most misleading way, generating really non-existent problems, like a EU approach to a "trustworthy AI", marked by human-like characteristics: 1. Human agency and oversight 2. Technical robustness and safety 3. Privacy and data governance 4. Transparency 5. Diversity, non-discrimination and fairness 6. Societal and environmental wellbeing 7. Accountability.
A major conceptual concern with today's AI is its extreme statistical reductionism, like "the regression to mediocrity", mixed with anthropocentrism and anthropomorphism, mimicking human brain or mind or behavior.
In fact, Real AI embraces philosophy, ontology, logic, semantics, linguistics, computer science, and mathematics, as well physical sciences, cognitive sciences, social sciences and technological science, with all engineering, be it electrical, electronic, mechanical, information, software, IT, and systems engineering, or biomedical, nano-, neuro-, quantum, and cognitive engineering. The RAI's world model looks as it is graphed here:
Global AI Academy Platform: Human Learning of Machine Intelligence and Learning
The Global AI Academy is an innovative knowledge digital platform to share the SOTA knowledge about the most advanced digital and emerging technologies. The GAIA to act through a global AI platform to spread worldwide the state-of-the art knowledge of the leading edge innovations, inventions, applications, and possible effects in the most critical fields of socio-technological change:
causal machine learning and robotics,
data science and big data,
cloud and edge computing,
the Internet of Thing and 5-6G,
cybersecurity,
digital reality,
smart cities,
intelligent nations,
smart sustainable world, i-world.
The GAIA is to give a real AI education, re-skilling and up-skilling narrow and weak AI mindsets. In a cutting-edge business perspective, it is advising how to set up competence centers (CC) or centers of excellence (COE) in real AI.
The idea of establishing a CC or COE in real AI is particularly radical. For most existing CoEs are largely about the weak and narrow AI of statistic ML and predictive data analytics, like as recently established by Deutsche Bank, J.P. Morgan Chase, Pfizer, Procter & Gamble, Anthem, and Farmers Insurance. Many tech strategists, visionaries and experts have poor understanding what kinds of AI CoEs the Real AI Economy is in need, AI Centers Of Excellence Accelerate AI Industry Adoption ; How to Set Up an AI Center of Excellence
The GAIA is to assist to create a smart data strategy with a vision for Real AI in the company, teaching executives to know what AI is, what it can do, and how it might enable new business models and strategies.
The current understanding of AI and ML is so poor that many businesses believe, like as ARK Invest , that the language models as the OpenAI’s?GPT-3 “understand” language, and "the statistic deep learning can create?more economic value?than the internet did".
Partly it could be explained with all sorts of Narrow AI and Statistic ML courses and programs, like ML by Stanford University or Machine Learning Andrew Ng Courses, misleading its students in a mass scale.
Why We All Need the Next Generation Machine Intelligence, Causal AI and Explainable ML
Today's Narrow and Weak AI of Machine Learning and Data Science is not a Real or True AI, be it large-scale language models as 17bn Turing-NLG, 175bn GPT-3, 1.75T Wu Dao 2.0, big tech ML platforms, recommending engines, digital assistants, self-driving transportation, or lethal autonomous weapon systems?(LAWS), autonomous weapon systems (AWS), robotic weapons, killer robots operating in the air, on land, on water, under water, or in space?
Its ML/DL algorithms and models are heavily relying on the statistical learning theory instead of causal learning, thus predicting spurious correlations instead of meaningful causation. This makes a critical difference for the whole enterprise, its applications, prospects, and impacts on every parts of human life.?
We have to be intelligently critical and fully objective as modern science demands it, as far as it concerns all of us and our human future. The AI world has been flooded with a series of gigantic language model projects promoted as the last word in AI. First, OpenAI shocked the world a year ago with GPT-3. In turn, Google presented LaMDA and MUM, two narrow AIs as revolutionizing chatbots and the search engine. And now the Beijing Academy of Artificial Intelligence (BAAI) conference presented Wu Dao 2.0.
All the language model applications heavily rely on ML/DL/AI technology which is fundamentally defective. This might be clear from the Turing award winners speeches, Hinton, LeCun and Bengio, who came out with their Deep Learning’s concerns, expressed at AAAI 2020.
Yann LeCun presented the top 3 challenges to Deep Learning:
Learning with fewer labeled samples
Learning to reason
Learning to plan complex action sequences
Geoff Hinton — "It's about the problems with CNNs and why they’re rubbish”:
CNNs not good at dealing with rotation or scaling
CNNs do not understand images in terms of objects and their parts
CNNs are brittle against adversarial examples
Yoshua Bengio — “Neural Networks need to develop consciousness”:
Should generalize faster from fewer examples
Learn better models from the world, like common sense
Get better at “System 2” thinking (slower, methodological thinking as opposed to fast recognition).
To make ML/DL true modules of AI, it is necessary to focus on 3 things:
1. Transfer or Generalization of Learning
2. Replacing Statistical Learning and Correlative Inferences with Causal Reasoning and Learning
3. Embedding Universal Concepts of the World discarding human-biased tabula rasa theory that individuals are born without built-in mental content and all knowledge comes from the sensory data of experience or perception.
Why machines are still far from having “the essence of intelligence”
In reality, there are no real AI systems, tools, libraries, and platforms in existence, be it Azure, Caffe, Tensorflow, Torch, IBM Watson, Keras, or what else.
This all is about advanced data analytics, predictive models, statistic learning, computational statistics and automation software, hyped as “AI, ML, or DL”.
Again, today's AI, in fact, a false and counterfeit AI. At the very best, it is a sort of automatic learning technologies, ML/DL/NN pattern-recognizers, which are essentially mathematical and statistical in nature, unable to act intuitively or model their environment, being with zero intelligence, nil learning and NO understanding. It is well sketched below how the ML model is usually built:
?Overall, there are a wide number of conceptions of machine intelligence, as sampled below:
1. "Understanding of the world,?Ability to interact with the world, and Self-awareness".
2. “There are degrees, or levels, of intelligence, and these are determined by: 1) the computational power of the system’s brain (or computer), 2) the sophistication of the algorithms the system uses for sensory processing, world modelling, behaviour generating, value judgement, and global communication, and 3) the information and values the system and stored in its memory.”?
3.?"Intelligence measures an agent’s ability to achieve goals in a wide range of environments" or the “the capacity to realize complex goals”.
Most of them are aligned with the today’s narrow and weak AI/ML/DL, strong AI or AGI, and superhuman AI, marked by implicit anthropocentric and anthropomorphic conceptions, ethical issues, and the pursuit of human-like intelligence as the golden standard for AI.?
As an example, the?OECD report ?(the Principles) aim to guide governments, organisations, and individuals to design and run human-like AI systems. In particular, the Principles include five key recommendations for the anthropocentric and anthropomorphic AI (AAAI):
领英推荐
This heavily affects the national AI strategies of EU Member States, Norway, and Switzerland.
The mainstream AI is missing the complexity, versatility and wholeness of intelligence, coming from its effective interactions, understanding, learning or computing the world itself, the reality with all its structures, phenomena, mechanisms and forms.
“The essence of intelligence” implies the power to understand reality, physical, mental, social, digital or virtual, to make causal predictions about basic aspects of it—to observe one thing and then use background knowledge to figure out what other things must also be true.?
Real AI vs. Climate Change and Global Warming
Real AI as a Causal Machine Intelligence and Learning, complementing human intelligence, is the only real way to save our world and humanity from the imminent peril coming from anthropogenic global risks, as climate change and global warming.
Climate change, with all its effects, as storms, droughts, fires, and flooding, is one of the greatest challenges facing humanity, and Real AI could be here a critical solution.
There are some high-tech efforts, but they involve ML/DL solutions having little to do with real world problems. I mean a team of ecologists and machine learning experts, who proposed "possible machine-learning interventions in 13 domains, from electricity systems to farms and forests to climate prediction. Within each domain, it breaks out the contributions for various subdisciplines within machine learning, including computer vision, natural-language processing, and reinforcement learning". Tackling Climate Change with Machine Learning
The GAIA focuses on A European Green Deal
Climate change and environmental degradation are an existential threat to Europe and the world. To overcome these challenges, the European Green Deal?will transform the EU into a modern, resource-efficient and competitive economy, ensuring:
The European Green Deal is also our lifeline out of the COVID-19 pandemic.?One third of the 1.8 trillion euro?investments from the NextGenerationEU Recovery Plan, and the EU’s seven-year budget will finance the European Green Deal.
Global Human-AI Platform as the Universal GPT (General-Purpose Technology)
The Real AI Science and Technology is emerging as the best of all the best ideas which changed and still changing the world:
The Global AI Platform implies a unifying model of reality/world in terms of causality/actuality, mentality/intelligence and computing/data /virtuality/cyberspace, where neural networks are brain-encoded causal networks.
Such as a prime, background and internally encoded knowledge serves as an integrative world model, WorldNet, the universal “master algorithm” that would unlock a Real/Global AI, integrating all sorts of MI and ML with human intelligence (HI):
HMIL = HI + Machine Intelligence [ANI, CC, AGI, ASI] + Machine Learning (DNNs) = Global AI = Real Intelligence
HMIL ?integrates all valuable approaches and techniques, such as ML (of which deep learning and reinforcement learning are specific examples), machine reasoning (which includes planning, scheduling, knowledge representation and reasoning, search, and optimization), and robotics (which includes control, perception, sensors and actuators, as well as the integration of all other techniques into cyber-physical systems).
Human intelligence is to be embraced by the global AI together with other emerging technologies and various sorts of AIs:
The Global or Real AI platform =
Human Intelligence,
Big Data Analytics,
Digital Reality,
Robotics,
Narrow AI, ML. DL,
AGI,
ASI,
the Internet of Everything.
Conclusion
The GAIA's mission is to transfer a critical learning worldwide that there are two broad types of Machine Intelligence, General [Scientific or Real] AI vs. Narrow {Statistic or Symbolic} AI, with all possible consequences for learning and education, cutting-edge technology development, business and policy and all sides of human life.
And the real/general/causal AI is emerging as the integral human-machine digital GPT of AI, ML, DL Robotics, and other emerging technologies, being all about reality, mentality and virtuality, or digital reality, relying on the scientific world models and substantial causality instead of spurious correlations and statistic models.
Sources
Everything You Know about Artificial Intelligence or Machine Intelligence and Learning is Old and Obsolete
What is real intelligence? What is natural intelligence and artificial intelligence and how are they different from each other?
WU DAO 2.0: WHY CHINA IS LEADING THE ARTIFICIAL INTELLIGENCE RACE?
HOW TO CREATE AN ARTIFICIAL INTELLIGENCE GENERAL TECHNOLOGY PLATFORM
MATHEMATICS OF MACHINE LEARNING: DATA, ALGORITHM, MODEL, AND CAUSAL LEARNING
WHAT IS THE DIFFERENCE BETWEEN THE LEARNING CURVE OF MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE?
Intelligent Governments and European Green Deal: Towards Smart Green Europe
The GAIA Books to Read:
Supplement: The Real AI and ML Knowledge Base for the GAIA
The Real AI and ML Knowledge Base for the GAIA is based on a dozen of original i-books, as above, hundreds of original articles, all added up with 2.2k answers given on the AI Quora Platform, viewed by 1.5 m readers and widely shared.
Some samples follow:
Statistical learning theory?is a framework for machine?learning?drawing from the fields of?statistics?and functional analysis.
Statistical learning theory?deals with the problem of finding a predictive function based on data.
This image represents your deep intelligence, being an example of overfitting in machine learning. The red dots represent training set data. The green line represents the true functional relationship, while the blue line shows the learned function, which has fallen victim to overfitting.
Classification is very common for machine learning applications. In facial recognition, for instance, a picture of a person's face would be the input, and the output label would be that person's name. The input would be represented by a large multidimensional vector whose elements represent pixels in the picture.
After learning a function based on the training set data, that function is validated on a test set of data, data that did not appear in the training set.
Statistical learning theory is the base of applications, as machine deep learning, including supervised learning, unsupervised learning, online learning, and reinforcement learning, computer vision, speech recognition, bioinformatics.
All what you need to know about ML/DL/NN contains in The Elements of Statistical Learning:?Data Mining, Inference, and Prediction.
Second Edition February 2009
So, the fathers of machine learning must be its authors:
Just note one big deep hidden truth:
Machine learning is impersonated as AI, becoming, in fact, a false and counterfeit or fakeAI. At the very best, it is a sort of automatic learning technologies, ML/DL/NN pattern-recognizers, which are essentially mathematical and statistical in nature, unable to act intuitively or model their environment, being with zero intelligence, nil learning and NO any understanding.
Not so far, within 5 years.
In fact, you are asking how to integrate AI into system software platform, as operating systems, to create superintelligent OSs.
It maybe operating systems like Apple's iOS, Google's Android, Microsoft's Windows Phone, BlackBerry's BlackBerry 10, Samsung's/Linux Foundation's Tizen and Jolla's Sailfish OS; macOS, GNU/Linux, computational science software, game engines, industrial automation, and software as a service applications.
It may be web browsers such as Internet Explorer, Chrome OS and Firefox OS for smartphones, tablet computers and smart TVs, cloud-based software or specialized classes of operating systems, such as embedded and real-time systems.
So, I can give a cue, while the full description is in a top classified, proprietary White Book having the basic specifications how to develop a real AI as a causal machine intelligence and learning:
Engineering Human-Machine Superintelligence by 2025: AI for Everything and Everyone (AI4EE)
WHITE PAPER | January 17, 2021 | EIS ENCYCLOPEDIC INTELLIGENCE SYSTEMS LTD
Here is an heuristic rule, each problem in science and technology is decided by adding up a new abstraction level.
ICT with computer science is not any exclusion.
All what we need is to innovate a sort of universal dataset as a classified system of all data, arranged in categories, classes, types and tokens.. It is designed as all data code layer over the universal character set code layer, or UNICODE, while applying ontology [describing concepts, relationships between entities, and categories of things].
All what you need is to create UniDataCode providing OPs, mobile or desktop, with embedded semantics to reason over data and operate with heterogeneous data sources.
For more info, search "Engineering a Symbiotic Superintelligence by 2025: meeting Musk's concerns for $100 billion".
Original Top Classified Resource
The author shares some original discoveries from a top classified, proprietary White Book having the basic specifications how to develop a real AI as a causal machine intelligence and learning:
Engineering Human-Machine Superintelligence by 2025: AI for Everything and Everyone (AI4EE)
WHITE PAPER | January 17, 2021 | EIS ENCYCLOPEDIC INTELLIGENCE SYSTEMS LTD