Lofti Zadeh, the 95-year-old mathematician, computer scientist, electrical engineer, artificial intelligence researcher, and professor emeritus of computer science at the University of California, Berkeley, US, did not know, at one point in time, much about subways. But, he knew the secret of trains, which purred through the outskirts of Sendai, Japan, halting with uncanny precision, and saving ten per cent of the average fuel used. Zadeh also apparently knew that the trains were working on a set of maxims that made up ‘fuzzy logic.’ This was, indeed, the magic word he had first ‘coined,’ in the mid-1960s — at a time when conventional logic was the rule, dividing as it did the world into yes and no, or black and white.
Zadeh’s proposition was classically different: the entire world is in shades of grey. It was something that, he also argued, could make computers ‘think’ like people. Curiously, the American academic community wasn’t impressed. It ridiculed the concept per se, and called fuzzy logic ‘futuristic fantasy.’ Not so the Japanese, who saw the logic of fuzzy logic — this was reason enough why companies like Matsushita and Sony began selling it back to the land of its origin. Today, fuzzy logic is being extensively used and incorporated in computer technology and a host of other engineering, medical and scientific processes, including high-tech gizmos, aside from household gadgets — making them almost as ‘smart’ as their owners.
Well, do we not use fuzzy logic to make housework easy? You toss your laundry into a washing machine and push a button — the machine does the rest. There is actually no fuss, no frill, no measuring the detergent, or choosing a wash temperature, or selecting a cycle. Also, think about using a microwave oven that ‘watches’ over meals, set to cook, with more sensitivity than a human being. In Japan, this has been used for a long time. This is also motivation enough why the technology is fast catching up with reality, what with products applying fuzzy logic now readily available in all markets.
Fuzzy logic is actually not logic that is fuzzy, but logic that describes and tames fuzziness. It is based on the concept that all things are a matter of degree. As a matter of fact, fuzziness underscores, perhaps, the most famous philosophical scheme ever devised: Plato’s theory of ideals. The great philosopher was aware of degrees of truth everywhere, and he often reconciled from them. For Plato, to use a simile, no chair was perfect; it was only a chair to a certain degree. Nothing in the world is perfect. Everything, the great man said, comes with similar grades of imperfection.
It is true that Plato’s ideals were not mere intellectual gems over practical sense. They had deep, primary effects. For example, they were able to circumvent or even overcome the popular notion of essentialism — one of the foremost obstacles to the theory of evolution. Though Plato did, at first, confuse all partial contradictions with the sum total, while viewing the harmony between objects as conflict between tall and short, he did see the ‘light’ and re-invent his philosophy. It was, of course, much later that he eliminated ‘fuzziness’ from existence and altered his perception of the world. This is not all. Closer home, Hinduism, or India’s rich, tradition of logic, still ‘holds’ on to the idea that the physical world is maya, or illusion. Likewise, Buddhism has distinctly fuzzy elements. The Buddha was, perhaps, the world’s first fuzzy theorist. His sayings turn the Law of Contradictions completely inside out.
Back to Zadeh. Complex systems, as he envisaged, trigger the idea of fuzzy sets. A chair, according to him, distils an array of objects into one central notion. Furniture, he says, summarises it even more broadly, as much as words which define every core concept that may have blurred bounds. He also explains that language is “a system for assigning atomic and composite labels, words, phrases, and sentences,” to fuzzy sets. Language is, quite simply, a vast shorthand — the outstanding instance of our ability to summarise.
Zadeh reckons that fuzzy logic could handle several complexities in a similar way. To illustrate a simple example: when members in a class grow, for instance, they eventually exceed human comprehension. When this happens, the brain subsequently responds by summarising the class into chunks, labelled with words. As Bart Kosko, another fuzzy scholar, puts it so logically, “What makes society turn is science, and the language of science is maths, and the structure of maths is logic, and the bedrock of logic is Aristotle, and that’s what goes out with fuzzy.”
How far can fuzzy logic go, you may well ask. According to Kosko, it has lead to, or will dramatically transform, present and futuristic scientific miracles. These include in a nutshell:
Vast expert decision-makers, who will theoretically be able to refine the wisdom of every document ever written
Smart cars with sonar devices that pump the brakes, if the car ahead stops too quickly. With a fuzzy navigator, a computerised map, emitters and receivers on asphalt or concrete roads, such vehicles could drive themselves
Robots with a human-like repertoire of behaviour; or, computers that understand, feel and respond to normal human language and emotions; or, machines that write prize-winning novels and screenplays in a selected style, such as Ernest Hemingway’s
Molecule-sized soldiers of health — nano-particles — that roam the bloodstream, killing cancer cells and/or slow down the aging process, among a host of other forays.
Fuzzy logic is already playing a pivotal role in medicine. To cull some exemplars — by courtesy of researchers, Angela Torres-Iglesias and Juan J Nieto — fuzzy logic crosses several disease groups to predict the response to treatment with citalopram — an anti-depressant drug — in alcohol dependence, analyse diabetic neuropathy [nerve damage] and detect early diabetic retinopathy, a complication that affects the eyes. It also helps to determine appropriate lithium [a psychiatric medication] dosage and calculate volumes of brain tissue from magnetic resonance imaging [MRI], besides analysing functional MRI data. Fuzzy logic helps to characterise stroke subtypes and/or co-existing causes of ischaemic stroke [obstruction within a blood vessel supplying blood to the brain]. It helps improve decision-making in radiation therapy, control hypertension [high blood pressure] during anaesthesia, determine flexor-tendon [tissues that help control movement in your hand] repair techniques, and detect breast, lung and prostate cancers. It, likewise, assists in the diagnosis of central nervous systems tumours and discriminate benign skin lesions from malignant melanomas [cancer]. Other uses include visualising nerve fibres in the human brain and exemplifying quantitative estimates of drug use, studying key components in schizophrenia, a mental disorder, and mapping fuzzy epidemics, while making quality decisions in nursing care and overcoming electroacupuncture [EA] accommodation. EA is a modified technique in acupuncture — it stimulates ‘acupoints’ with electrical current, instead of manual manipulation.
You get the point. If fuzzy information can lead to rapid and logical progress, there’s no reason why fuzzy systems, or feasible fuzzy devices, should not be more and more commonplace. To pick some fuzzy examples — insulin pumps, incubators for premature infants, advanced devices to induce labour by smoothening the flow of anaesthetics and drugs to the expectant mother, pool chlorinators, throttle controllers for racing powerboats to keep their bows from pitching out of the water, aquarium management systems, heart pacemakers, light dimmers, liquid cooling systems for computer workstations, and a host of other applications. The list is, of course, incomplete.
To summarise and quote technologists Daniel McNeill and Paul Freiberger, authors of Fuzzy Logic, a classy book, for what would make fuzzy logic get bigger: “It [fuzzy logic] forsakes not precision, but pointless precision. It abandons an either or hairline that never existed, and brightens technology at the cost of a tiny blur. It is neither a ‘dream’ like artificial intelligence, nor a dead-end, a little trick for washers and cameras. It is here today, and no matter what the brand name on the label, it will probably be here tomorrow.” If this isn’t fuzzy logic exemplified, validated, replicated and/or proved, what is?