Ten years ago, if you mentioned the term “artificial intelligence” in a boardroom there’s a good chance you would have been laughed at. For most people it would bring to mind sentient, sci-fi machines such as 2001: A Space Odyssey’s HAL or Star Trek’s Data.
Today it is one of the hottest buzzwords in business and industry. AI technology is a crucial lynchpin of much of the digital transformation taking place today as organisations position themselves to capitalize on the ever-growing amount of data being generated and collected.
So how has this change come about? Well partly it is due to the Big Data revolution itself. The glut of data has led to intensified research into ways it can be processed, analysed and acted upon. Machines being far better suited to humans than this work, the focus was on training machines to do this in as “smart” a way as is possible.
This increased interest in research in the field – in academia, industry and among the open source community which sits in the middle – has led to breakthroughs and advances that are showing their potential to generate tremendous change. From healthcare to self-driving cars to predicting the outcome of legal cases, no one is laughing now!
The concept of what defines AI has changed over time, but at the core there has always been the idea of building machines which are capable of thinking like humans.
After all, human beings have proven uniquely capable of interpreting the world around us and using the information we pick up to effect change. If we want to build machines to help us to this more efficiently, then it makes sense to use ourselves as a blueprint!
AI, then, can be thought of as simulating the capacity for abstract, creative, deductive thought – and particularly the ability to learn – using the digital, binary logic of computers.
Research and development work in AI is split between two branches. One is labelled “applied AI” which uses these principles of simulating human thought to carry out one specific task. The other is known as “generalised AI” – which seeks to develop machine intelligences that can turn their hands to any task, much like a person.
Research into applied, specialised AI is already providing breakthroughs in fields of study from quantum physics where it is used to model and predict the behaviour of systems comprised of billions of subatomic particles, to medicine where it being used to diagnose patients based on genomic data.
In industry, it is employed in the financial world for uses ranging from fraud detection to improving customer service by predicting what services customers will need. In manufacturing it is used to manage workforces and production processes as well as for predicting faults before they occur, therefore enabling predictive maintenance.
In the consumer world more and more of the technology we are adopting into our everyday lives is becoming powered by AI – from smartphone assistants like Apple’s Siri and Google’s Google Assistant, to self-driving and autonomous cars which many are predicting will outnumber manually driven cars within our lifetimes.
Generalised AI is a bit further off – to carry out a complete simulation of the human brain would require both a more complete understanding of the organ than we currently have, and more computing power than is commonly available to researchers. But that may not be the case for long, given the speed with which computer technology is evolving. A new generation of computer chip technology known as neuromorphic processors are being designed to more efficiently run brain-simulator code. And systems such as IBM’s Watson cognitive computing platform use high-level simulations of human neurological processes to carry out an ever-growing range of tasks without being specifically taught how to do them.
All of these advances have been made possible due to the focus on imitating human thought processes. The field of research which has been most fruitful in recent years is what has become known as “machine learning”. In fact, it’s become so integral to contemporary AI that the terms “artificial intelligence” and “machine learning” are sometimes used interchangeably.
However, this is an imprecise use of language, and the best way to think of it is that machine learning represents the current state-of-the-art in the wider field of AI. The foundation of machine learning is that rather than have to be taught to do everything step by step, machines, if they can be programmed to think like us, can learn to work by observing, classifying and learning from its mistakes, just like we do.
The application of neuroscience to IT system architecture has led to the development of artificial neural networks– and although work in this field has evolved over the last half century it is only recently that computers with adequate power have been available to make the task a day-to-day reality for anyone except those with access to the most expensive, specialised tools.
Perhaps the single biggest enabling factor has been the explosion of data which has been unleashed since mainstream society merged itself with the digital world. This availability of data – from things we share on social media to machine data generated by connected industrial machinery – means computers now have a universe of information available to them, to help them learn more efficiently and make better decisions.
Even if robots don’t eradicate us or turn us into living batteries, a less dramatic but still nightmarish scenario is that automation of labour (mental as well as physical) will lead to profound societal change – perhaps for the better, or perhaps for the worse.
In
computer science,
artificial intelligence (
AI), sometimes called
machine intelligence, is
intelligence demonstrated by
machines, in contrast to the
natural intelligence displayed by humans. Leading AI textbooks define the field as the study of "
intelligent agents": any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals.
[1] Colloquially, the term "artificial intelligence" is often used to describe machines (or computers) that mimic "cognitive" functions that humans associate with the
human mind, such as "learning" and "problem solving".
Artificial intelligence can be classified into three different types of systems:
- Analytical
- Human-inspired
- Humanized artificial intelligence.[8]
Analytical AI has only characteristics consistent with
cognitive intelligence; generating a cognitive representation of the world and using learning based on past experience to inform future decisions. Human-inspired AI has elements from cognitive and
emotional intelligence; understanding human emotions, in addition to cognitive elements, and considering them in their
decision making. Humanized AI shows characteristics of all types of competencies (i.e., cognitive, emotional, and
social intelligence), is able to be
self-conscious and is
self-aware in interactions.
Artificial intelligence was founded as an academic discipline in 1956, and in the years since has experienced several waves of optimism,
[9][10] followed by disappointment and the loss of funding (known as an "
AI winter"),
[11][12] followed by new approaches, success and renewed funding.
[10][13] For most of its history, AI research has been divided into subfields that often fail to communicate with each other.
[14] These sub-fields are based on technical considerations, such as particular goals (e.g. "
robotics" or "machine learning"),
[15] the use of particular tools ("logic" or
artificial neural networks), or deep philosophical differences.
[16][17][18] Subfields have also been based on social factors (particular institutions or the work of particular researchers).
[14]
The traditional problems (or goals) of AI research include
reasoning,
knowledge representation,
planning,
learning,
natural language processing,
perception and the ability to move and manipulate objects.
[15] General intelligence is among the field's long-term goals.
[19] Approaches include
statistical methods,
computational intelligence, and
traditional symbolic AI. Many tools are used in AI, including versions of
search and mathematical optimization,
artificial neural networks, and
methods based on statistics, probability and economics. The AI field draws upon
computer science,
information engineering,
mathematics,
psychology,
linguistics,
philosophy, and many other fields.
The field was founded on the assumption that
human intelligence "can be so precisely described that a machine can be made to simulate it".
[20] This raises philosophical arguments about the nature of the
mind and the ethics of creating artificial beings endowed with human-like intelligence. These issues have been explored by
myth,
fiction and
philosophy since
antiquity.
[21] Some people also consider AI to be
a danger to humanity if it progresses unabated.
[22] Others believe that AI, unlike previous technological revolutions, will create a
risk of mass unemployment.
[23]
The study of mechanical or
"formal" reasoning began with
philosophers and mathematicians in antiquity. The study of mathematical logic led directly to
Alan Turing's
theory of computation, which suggested that a machine, by shuffling symbols as simple as "0" and "1", could simulate any conceivable act of mathematical deduction. This insight, that digital computers can simulate any process of formal reasoning, is known as the
Church–Turing thesis.
[27] Along with concurrent discoveries in
neurobiology,
information theory and
cybernetics, this led researchers to consider the possibility of building an electronic brain. Turing proposed changing the question from whether a machine was intelligent, to "whether or not it is possible for machinery to show intelligent behaviour".
[28] The first work that is now generally recognized as AI was
McCullouch and
Pitts' 1943 formal design for
Turing-complete "artificial neurons".
The field of AI research was born at
a workshop at
Dartmouth College in 1956.
[30] Attendees
Allen Newell (
CMU),
Herbert Simon (CMU),
John McCarthy (
MIT),
Marvin Minsky (MIT) and
Arthur Samuel (
IBM) became the founders and leaders of AI research.
[31] They and their students produced programs that the press described as "astonishing": computers were learning
checkers strategies (c. 1954)
[33] (and by 1959 were reportedly playing better than the average human),
[34] solving word problems in algebra, proving
logical theorems (
Logic Theorist, first run c. 1956) and speaking English.
[35] By the middle of the 1960s, research in the U.S. was heavily funded by the
Department of Defense[36] and laboratories had been established around the world.
[37] AI's founders were optimistic about the future:
Herbert Simon predicted, "machines will be capable, within twenty years, of doing any work a man can do".
Marvin Minsky agreed, writing, "within a generation ... the problem of creating 'artificial intelligence' will substantially be solved".
[9]
They failed to recognize the difficulty of some of the remaining tasks. Progress slowed and in 1974, in response to the criticism of
Sir James Lighthill and ongoing pressure from the US Congress to fund more productive projects, both the U.S. and British governments cut off exploratory research in AI. The next few years would later be called an "
AI winter",
[11] a period when obtaining funding for AI projects was difficult.
In the early 1980s, AI research was revived by the commercial success of
expert systems,
[39] a form of AI program that simulated the knowledge and analytical skills of human experts. By 1985, the market for AI had reached over a billion dollars. At the same time, Japan's
fifth generation computer project inspired the U.S and British governments to restore funding for
academic research.
[10] However, beginning with the collapse of the
Lisp Machine market in 1987, AI once again fell into disrepute, and a second, longer-lasting hiatus began.
[12]
In the late 1990s and early 21st century, AI began to be used for logistics,
data mining,
medical diagnosis and other areas.
[24] The success was due to increasing computational power (see
Moore's law), greater emphasis on solving specific problems, new ties between AI and other fields (such as
statistics,
economics and
mathematics), and a commitment by researchers to mathematical methods and scientific standards.
[40] Deep Blue became the first computer chess-playing system to beat a reigning world chess champion,
Garry Kasparov, on 11 May 1997.
In 2011, a
Jeopardy! quiz show exhibition match,
IBM's
question answering system,
Watson, defeated the two greatest
Jeopardy! champions,
Brad Rutter and
Ken Jennings, by a significant margin.
Faster computers, algorithmic improvements, and access to
large amounts of data enabled advances in
machine learning and perception; data-hungry
deep learning methods started to dominate accuracy benchmarks
around 2012.
[43] The
Kinect, which provides a 3D body–motion interface for the
Xbox 360 and the
Xbox One, uses algorithms that emerged from lengthy AI research
[44] as do
intelligent personal assistants in
smartphones.
[45] In March 2016,
AlphaGo won 4 out of 5 games of
Go in a match with Go champion
Lee Sedol, becoming the first
computer Go-playing system to beat a professional Go player without
handicaps.
[7][46] In the 2017
Future of Go Summit,
AlphaGo won a
three-game match with
Ke Jie,
[47] who at the time continuously held the world No. 1 ranking for two years.
[48][49] This marked the completion of a significant milestone in the development of Artificial Intelligence as Go is a relatively complex game, more so than Chess.
According to
Bloomberg's Jack Clark, 2015 was a landmark year for artificial intelligence, with the number of software projects that use AI
Google increased from a "sporadic usage" in 2012 to more than 2,700 projects. Clark also presents factual data indicating the improvements of AI since 2012 supported by lower error rates in image processing tasks.
[50] He attributes this to an increase in affordable
neural networks, due to a rise in cloud computing infrastructure and to an increase in research tools and datasets.
[13] Other cited examples include Microsoft's development of a Skype system that can automatically translate from one language to another and Facebook's system that can describe images to blind people.
[50] In a 2017 survey, one in five companies reported they had "incorporated AI in some offerings or processes".
[51][52] Around 2016,
China greatly accelerated its government funding; given its large supply of data and its rapidly increasing research output, some observers believe it may be on track to becoming an "AI superpower".
[53][54] However, it has been acknowledged that reports regarding artificial intelligence have tended to be exaggerated.
[55][56][57]
Definitions
Computer science defines AI research as the study of "
intelligent agents": any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals.
[1] A more elaborate definition characterizes AI as “a system’s ability to correctly interpret external data, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation.”
[58]
Basics
A typical AI analyzes its environment and takes actions that maximize its chance of success.
[1] An AI's intended
utility function (or goal) can be simple ("1 if the AI wins a game of
Go, 0 otherwise") or complex ("Do mathematically similar actions to the ones succeeded in the past"). Goals can be explicitly defined, or induced. If the AI is programmed for "
reinforcement learning", goals can be implicitly induced by rewarding some types of behavior or punishing others.
[a] Alternatively, an evolutionary system can induce goals by using a "
fitness function" to mutate and preferentially replicate high-scoring AI systems, similarly to how animals evolved to innately desire certain goals such as finding food. Some AI systems, such as nearest-neighbor, instead of reason by analogy, these systems are not generally given goals, except to the degree that goals are implicit in their training data. Such systems can still be benchmarked if the non-goal system is framed as a system whose "goal" is to successfully accomplish its narrow classification task.
[61]
AI often revolves around the use of
algorithms. An algorithm is a set of unambiguous instructions that a mechanical computer can execute.
[b] A complex algorithm is often built on top of other, simpler, algorithms. A simple example of an algorithm is the following (optimal for first player) recipe for play at
tic-tac-toe:
- If someone has a "threat" (that is, two in a row), take the remaining square. Otherwise,
- if a move "forks" to create two threats at once, play that move. Otherwise,
- take the center square if it is free. Otherwise,
- if your opponent has played in a corner, take the opposite corner. Otherwise,
- take an empty corner if one exists. Otherwise,
- take any empty square.
Many AI algorithms are capable of learning from data; they can enhance themselves by learning new
heuristics (strategies, or "rules of thumb", that have worked well in the past), or can themselves write other algorithms. Some of the "learners" described below, including Bayesian networks, decision trees, and nearest-neighbor, could theoretically, (given infinite data, time, and memory) learn to approximate any
function, including which combination of mathematical functions would best describe the world
[citation needed]. These learners could therefore, derive all possible knowledge, by considering every possible hypothesis and matching them against the data. In practice, it is almost never possible to consider every possibility, because of the phenomenon of "
combinatorial explosion", where the amount of time needed to solve a problem grows exponentially. Much of AI research involves figuring out how to identify and avoid considering broad range of possibilities that are unlikely to be beneficial.
[63] For example, when viewing a map and looking for the shortest driving route from
Denver to
New York in the East, one can in most cases skip looking at any path through
San Francisco or other areas far to the West; thus, an AI wielding a pathfinding algorithm like
A* can avoid the combinatorial explosion that would ensue if every possible route had to be ponderously considered in turn.
[65]
The earliest (and easiest to understand) approach to AI was symbolism (such as formal logic): "If an otherwise healthy adult has a fever, then they may have
influenza". A second, more general, approach is
Bayesian inference: "If the current patient has a fever, adjust the probability they have influenza in such-and-such way". The third major approach, extremely popular in routine business AI applications, are analogizers such as
SVM and
nearest-neighbor: "After examining the records of known past patients whose temperature, symptoms, age, and other factors mostly match the current patient, X% of those patients turned out to have influenza". A fourth approach is harder to intuitively understand, but is inspired by how the brain's machinery works: the
artificial neural network approach uses artificial "
neurons" that can learn by comparing itself to the desired output and altering the strengths of the connections between its internal neurons to "reinforce" connections that seemed to be useful. These four main approaches can overlap with each other and with evolutionary systems; for example, neural nets can learn to make inferences, to generalize, and to make analogies. Some systems implicitly or explicitly use multiple of these approaches, alongside many other AI and non-AI algorithms;
[66] the best approach is often different depending on the problem.
[68]
Learning algorithms work on the basis that strategies, algorithms, and inferences that worked well in the past are likely to continue working well in the future. These inferences can be obvious, such as "since the sun rose every morning for the last 10,000 days, it will probably rise tomorrow morning as well". They can be nuanced, such as "X% of
families have geographically separate species with color variants, so there is an Y% chance that undiscovered
black swans exist". Learners also work on the basis of "
Occam's razor": The simplest theory that explains the data is the likeliest. Therefore, according to Occam's razor principle, a learner must be designed such that it prefers simpler theories to complex theories, except in cases where the complex theory is proven substantially better.
The blue line could be an example of overfitting a linear function due to random noise.
Settling on a bad, overly complex theory gerrymandered to fit all the past training data is known as overfitting. Many systems attempt to reduce overfitting by rewarding a theory in accordance with how well it fits the data, but penalizing the theory in accordance with how complex the theory is. Besides classic overfitting, learners can also disappoint by "learning the wrong lesson". A toy example is that an image classifier trained only on pictures of brown horses and black cats might conclude that all brown patches are likely to be horses. A real-world example is that, unlike humans, current image classifiers don't determine the spatial relationship between components of the picture; instead, they learn abstract patterns of pixels that humans are oblivious to, but that linearly correlate with images of certain types of real objects. Faintly superimposing such a pattern on a legitimate image results in an "adversarial" image that the system misclassifies.[c][71][72][73]
Compared with humans, existing AI lacks several features of human "commonsense reasoning"; most notably, humans have powerful mechanisms for reasoning about "naïve physics" such as space, time, and physical interactions. This enables even young children to easily make inferences like "If I roll this pen off a table, it will fall on the floor". Humans also have a powerful mechanism of "folk psychology" that helps them to interpret natural-language sentences such as "The city councilmen refused the demonstrators a permit because they advocated violence". (A generic AI has difficulty discerning whether the ones alleged to be advocating violence are the councilmen or the demonstrators.)[76][77][78] This lack of "common knowledge" means that AI often makes different mistakes than humans make, in ways that can seem incomprehensible. For example, existing self-driving cars cannot reason about the location nor the intentions of pedestrians in the exact way that humans do, and instead must use non-human modes of reasoning to avoid accidents.[79][80][81]
Challenges of AI
The overall research goal of artificial intelligence is to create technology that allows computers and machines to function in an intelligent manner. The general problem of simulating (or creating) intelligence has been broken down into sub-problems. These consist of particular traits or capabilities that researchers expect an intelligent system to display. The traits described below have received the most attention.
[15]
Reasoning, problem solving
Early researchers developed algorithms that imitated step-by-step reasoning that humans use when they solve puzzles or make logical deductions.
[82] By the late 1980s and 1990s, AI research had developed methods for dealing with
uncertain or incomplete information, employing concepts from
probability and
economics.
[83]
These algorithms proved to be insufficient for solving large reasoning problems, because they experienced a "combinatorial explosion": they became exponentially slower as the problems grew larger.
[63] In fact, even humans rarely use the step-by-step deduction that early AI research was able to model. They solve most of their problems using fast, intuitive judgements.
[84]
An ontology represents knowledge as a set of concepts within a domain and the relationships between those concepts.
Knowledge representation
Knowledge representation[85] and
knowledge engineering[86] are central to classical AI research. Some "expert systems" attempt to gather together explicit knowledge possessed by experts in some narrow domain. In addition, some projects attempt to gather the "commonsense knowledge" known to the average person into a database containing extensive knowledge about the world. Among the things a comprehensive commonsense knowledge base would contain are: objects, properties, categories and relations between objects;
[87] situations, events, states and time;
[88] causes and effects;
[89] knowledge about knowledge (what we know about what other people know);
[90] and many other, less well researched domains. A representation of "what exists" is an
ontology: the set of objects, relations, concepts, and properties formally described so that software agents can interpret them. The
semantics of these are captured as
description logic concepts, roles, and individuals, and typically implemented as classes, properties, and individuals in the
Web Ontology Language.
[91] The most general ontologies are called
upper ontologies, which attempt to provide a foundation for all other knowledge
[92] by acting as mediators between
domain ontologies that cover specific knowledge about a particular knowledge domain (field of interest or area of concern). Such formal knowledge representations can be used in content-based indexing and retrieval,
[93] scene interpretation,
[94] clinical decision support,
[95] knowledge discovery (mining "interesting" and actionable inferences from large databases),
[96] and other areas.
[97]
Among the most difficult problems in knowledge representation are:
- Default reasoning and the qualification problem
- Many of the things people know take the form of "working assumptions". For example, if a bird comes up in conversation, people typically picture an animal that is fist-sized, sings, and flies. None of these things are true about all birds. John McCarthy identified this problem in 1969[98] as the qualification problem: for any commonsense rule that AI researchers care to represent, there tend to be a huge number of exceptions. Almost nothing is simply true or false in the way that abstract logic requires. AI research has explored a number of solutions to this problem.[99]
- The breadth of commonsense knowledge
- The number of atomic facts that the average person knows is very large. Research projects that attempt to build a complete knowledge base of commonsense knowledge (e.g., Cyc) require enormous amounts of laborious ontological engineering—they must be built, by hand, one complicated concept at a time.[100]
- The subsymbolic form of some commonsense knowledge
- Much of what people know is not represented as "facts" or "statements" that they could express verbally. For example, a chess master will avoid a particular chess position because it "feels too exposed" or an art critic can take one look at a statue and realize that it is a fake. These are non-conscious and sub-symbolic intuitions or tendencies in the human brain.[103] Knowledge like this informs, supports and provides a context for symbolic, conscious knowledge. As with the related problem of sub-symbolic reasoning, it is hoped that situated AI, computational intelligence, or statistical AI will provide ways to represent this kind of knowledge.[103]
Planning
Intelligent agents must be able to set goals and achieve them.
[104] They need a way to visualize the future—a representation of the state of the world and be able to make predictions about how their actions will change it—and be able to make choices that maximize the
utility (or "value") of available choices.
[105]
In classical planning problems, the agent can assume that it is the only system acting in the world, allowing the agent to be certain of the consequences of its actions.
[106] However, if the agent is not the only actor, then it requires that the agent can reason under uncertainty. This calls for an agent that can not only assess its environment and make predictions, but also evaluate its predictions and adapt based on its assessment.
[107]
Learning
Machine learning (ML), a fundamental concept of AI research since the field's inception,
[109] is the study of computer algorithms that improve automatically through experience.
[110][111]
Unsupervised learning is the ability to find patterns in a stream of input, without requiring a human to label the inputs first.
Supervised learning includes both
classification and numerical
regression, which requires a human to label the input data first. Classification is used to determine what category something belongs in, and occurs after a program sees a number of examples of things from several categories. Regression is the attempt to produce a function that describes the relationship between inputs and outputs and predicts how the outputs should change as the inputs change.
[111] Both classifiers and regression learners can be viewed as "function approximators" trying to learn an unknown (possibly implicit) function; for example, a spam classifier can be viewed as learning a function that maps from the text of an email to one of two categories, "spam" or "not spam".
Computational learning theory can assess learners by
computational complexity, by
sample complexity (how much data is required), or by other notions of
optimization.
[112] In
reinforcement learning[113] the agent is rewarded for good responses and punished for bad ones. The agent uses this sequence of rewards and punishments to form a strategy for operating in its problem space.
Natural language processing
Natural language processing[114] (NLP) gives machines the ability to read and understand human language. A sufficiently powerful natural language processing system would enable natural-language user interfaces and the acquisition of knowledge directly from human-written sources, such as newswire texts. Some straightforward applications of natural language processing include information retrieval, text mining, question answering[115] and machine translation.[116] Many current approaches use word co-occurrence frequencies to construct syntactic representations of text. "Keyword spotting" strategies for search are popular and scalable but dumb; a search query for "dog" might only match documents with the literal word "dog" and miss a document with the word "poodle". "Lexical affinity" strategies use the occurrence of words such as "accident" to assess the sentiment of a document. Modern statistical NLP approaches can combine all these strategies as well as others, and often achieve acceptable accuracy at the page or paragraph level, but continue to lack the semantic understanding required to classify isolated sentences well. Besides the usual difficulties with encoding semantic commonsense knowledge, existing semantic NLP sometimes scales too poorly to be viable in business applications. Beyond semantic NLP, the ultimate goal of "narrative" NLP is to embody a full understanding of commonsense reasoning.[117]
Perception
Machine perception[118] is the ability to use input from sensors (such as cameras (visible spectrum or infrared), microphones, wireless signals, and active
lidar, sonar, radar, and
tactile sensors) to deduce aspects of the world. Applications include
speech recognition,
[119] facial recognition, and
object recognition.
[120] Computer vision is the ability to analyze visual input. Such input is usually ambiguous; a giant, fifty-meter-tall pedestrian far away may produce exactly the same pixels as a nearby normal-sized pedestrian, requiring the AI to judge the relative likelihood and reasonableness of different interpretations, for example by using its "object model" to assess that fifty-meter pedestrians do not exist.
[121]
Motion and manipulation
AI is heavily used in
robotics.
[122] Advanced
robotic arms and other
industrial robots, widely used in modern factories, can learn from experience how to move efficiently despite the presence of friction and gear slippage.
[123] A modern mobile robot, when given a small, static, and visible environment, can easily determine its location and
map its environment; however, dynamic environments, such as (in
endoscopy) the interior of a patient's breathing body, pose a greater challenge.
Motion planning is the process of breaking down a movement task into "primitives" such as individual joint movements. Such movement often involves compliant motion, a process where movement requires maintaining physical contact with an object.
[125][126] Moravec's paradox generalizes that low-level sensorimotor skills that humans take for granted are, counterintuitively, difficult to program into a robot; the paradox is named after
Hans Moravec, who stated in 1988 that "it is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility".
[127][128] This is attributed to the fact that, unlike checkers, physical dexterity has been a direct target of
natural selection for millions of years.
[129]
Social intelligence
Moravec's paradox can be extended to many forms of social intelligence.
[131][132] Distributed multi-agent coordination of autonomous vehicles remains a difficult problem.
[133] Affective computing is an interdisciplinary umbrella that comprises systems which recognize, interpret, process, or simulate human
affects. Moderate successes related to affective computing include textual
sentiment analysis and, more recently, multimodal affect analysis (see
multimodal sentiment analysis), wherein AI classifies the affects displayed by a videotaped subject.
[137]
In the long run, social skills and an understanding of human emotion and
game theory would be valuable to a social agent. Being able to predict the actions of others by understanding their motives and emotional states would allow an agent to make better decisions. Some computer systems mimic human emotion and expressions to appear more sensitive to the emotional dynamics of human interaction, or to otherwise facilitate
human–computer interaction.
[138] Similarly, some
virtual assistants are programmed to speak conversationally or even to banter humorously; this tends to give naïve users an unrealistic conception of how intelligent existing computer agents actually are.
[139]
General intelligence
Historically, projects such as the Cyc knowledge base (1984–) and the massive Japanese
Fifth Generation Computer Systems initiative (1982–1992) attempted to cover the breadth of human cognition. These early projects failed to escape the limitations of non-quantitative symbolic logic models and, in retrospect, greatly underestimated the difficulty of cross-domain AI. Nowadays, the vast majority of current AI researchers work instead on tractable "narrow AI" applications (such as medical diagnosis or automobile navigation).
[140] Many researchers predict that such "narrow AI" work in different individual domains will eventually be incorporated into a machine with
artificial general intelligence (AGI), combining most of the narrow skills mentioned in this article and at some point even exceeding human ability in most or all these areas.
[19][141] Many advances have general, cross-domain significance. One high-profile example is that
DeepMind in the 2010s developed a "generalized artificial intelligence" that could learn many diverse
Atari games on its own, and later developed a variant of the system which succeeds at
sequential learning.
[142][143][144] Besides
transfer learning,
[145] hypothetical AGI breakthroughs could include the development of reflective architectures that can engage in decision-theoretic metareasoning, and figuring out how to "slurp up" a comprehensive knowledge base from the entire unstructured
Web. Some argue that some kind of (currently-undiscovered) conceptually straightforward, but mathematically difficult, "Master Algorithm" could lead to AGI. Finally, a few "emergent" approaches look to simulating human intelligence extremely closely, and believe that
anthropomorphic features like an
artificial brain or simulated
child development may someday reach a critical point where general intelligence emerges.
[147][148]
Many of the problems in this article may also require general intelligence, if machines are to solve the problems as well as people do. For example, even specific straightforward tasks, like
machine translation, require that a machine read and write in both languages (
NLP), follow the author's argument (
reason), know what is being talked about (
knowledge), and faithfully reproduce the author's original intent (
social intelligence). A problem like machine translation is considered "
AI-complete", because all of these problems need to be solved simultaneously in order to reach human-level machine performance.
Approaches
There is no established unifying theory or
paradigm that guides AI research. Researchers disagree about many issues.
[149] A few of the most long standing questions that have remained unanswered are these: should artificial intelligence simulate natural intelligence by studying
psychology or
neurobiology? Or is
human biology as irrelevant to AI research as bird biology is to
aeronautical engineering?
[16] Can intelligent behavior be described using simple, elegant principles (such as
logic or
optimization)? Or does it necessarily require solving a large number of completely unrelated problems?
[17]
Cybernetics and brain simulation
Symbolic
When access to digital computers became possible in the mid 1950s, AI research began to explore the possibility that human intelligence could be reduced to symbol manipulation. The research was centered in three institutions:
Carnegie Mellon University,
Stanford and
MIT, and as described below, each one developed its own style of research.
John Haugeland named these symbolic approaches to AI "good old fashioned AI" or "
GOFAI".
[151] During the 1960s, symbolic approaches had achieved great success at simulating high-level "thinking" in small demonstration programs. Approaches based on
cybernetics or
artificial neural networks were abandoned or pushed into the background.
[152] Researchers in the 1960s and the 1970s were convinced that symbolic approaches would eventually succeed in creating a machine with
artificial general intelligence and considered this the goal of their field.
Cognitive simulation
Logic-based
Anti-logic or scruffy
Knowledge-based
When computers with large memories became available around 1970, researchers from all three traditions began to build
knowledge into AI applications.
[159] This "knowledge revolution" led to the development and deployment of
expert systems (introduced by
Edward Feigenbaum), the first truly successful form of AI software.
[39] A key component of the system architecture for all expert systems is the knowledge base, which stores facts and rules that illustrate AI.
[160] The knowledge revolution was also driven by the realization that enormous amounts of knowledge would be required by many simple AI applications.
Sub-symbolic
By the 1980s, progress in symbolic AI seemed to stall and many believed that symbolic systems would never be able to imitate all the processes of human cognition, especially
perception,
robotics,
learning and
pattern recognition. A number of researchers began to look into "sub-symbolic" approaches to specific AI problems.
[18] Sub-symbolic methods manage to approach intelligence without specific representations of knowledge.
Embodied intelligence
Within
developmental robotics, developmental learning approaches are elaborated upon to allow robots to accumulate repertoires of novel skills through autonomous self-exploration, social interaction with human teachers, and the use of guidance mechanisms (active learning, maturation, motor synergies, etc.).
[162][163]
Computational intelligence and soft computing
Statistical learning
Much of traditional
GOFAI got bogged down on
ad hoc patches to
symbolic computation that worked on their own toy models but failed to generalize to real-world results. However, around the 1990s, AI researchers adopted sophisticated mathematical tools, such as
hidden Markov models (HMM),
information theory, and normative Bayesian
decision theory to compare or to unify competing architectures. The shared mathematical language permitted a high level of collaboration with more established fields (like
mathematics, economics or
operations research).
[d] Compared with GOFAI, new "statistical learning" techniques such as HMM and neural networks were gaining higher levels of accuracy in many practical domains such as
data mining, without necessarily acquiring a semantic understanding of the datasets. The increased successes with real-world data led to increasing emphasis on comparing different approaches against shared test data to see which approach performed best in a broader context than that provided by idiosyncratic toy models; AI research was becoming more
scientific. Nowadays results of experiments are often rigorously measurable, and are sometimes (with difficulty) reproducible.
[40][168] Different statistical learning techniques have different limitations; for example, basic HMM cannot model the infinite possible combinations of natural language. Critics note that the shift from GOFAI to statistical learning is often also a shift away from
explainable AI. In AGI research, some scholars caution against over-reliance on statistical learning, and argue that continuing research into GOFAI will still be necessary to attain general intelligence.
Integrating the approaches
- Intelligent agent paradigm
- An intelligent agent is a system that perceives its environment and takes actions which maximize its chances of success. The simplest intelligent agents are programs that solve specific problems. More complicated agents include human beings and organizations of human beings (such as firms). The paradigm allows researchers to directly compare or even combine different approaches to isolated problems, by asking which agent is best at maximizing a given "goal function". An agent that solves a specific problem can use any approach that works—some agents are symbolic and logical, some are sub-symbolic artificial neural networks and others may use new approaches. The paradigm also gives researchers a common language to communicate with other fields—such as decision theory and economics—that also use concepts of abstract agents. Building a complete agent requires researchers to address realistic problems of integration; for example, because sensory systems give uncertain information about the environment, planning systems must be able to function in the presence of uncertainty. The intelligent agent paradigm became widely accepted during the 1990s.[172]
- Agent architectures and cognitive architectures
- Researchers have designed systems to build intelligent systems out of interacting intelligent agents in a multi-agent system.[173] A hierarchical control system provides a bridge between sub-symbolic AI at its lowest, reactive levels and traditional symbolic AI at its highest levels, where relaxed time constraints permit planning and world modelling.[174] Some cognitive architectures are custom-built to solve a narrow problem; others, such as Soar, are designed to mimic human cognition and to provide insight into general intelligence. Modern extensions of Soar are hybrid intelligent systems that include both symbolic and sub-symbolic components.[175][176][177]
