Heuristic Research in Hong Kong

By Dr. Juan Carlos Olabe-Basogain
Dr. Juan Carlos Olabe (Electrical & Computer Engineering) presented a paper entitled “Solving Complex Problems with a Computational Mind: An Alternative to Heuristic Search” at the 2015 Hong Kong International Conference on Education in late November. In addition, Dr. Olabe and members of his research team met with leaders of the Lasallian family of schools in Hong Kong (there are eight Lasallian K-12 schools in the city). Two areas of collaboration were defined: the integration of the Computational Thinking curriculum developed by Dr. Olabe’s team in the local Lasallian schools, and the creation of a Computational Thinking Institute at La Salle College in Hong Kong to provide academic resources and teaching training support to the Lasallian schools in Asia and the Pacific Region as well as other school systems in the region. Hong Kong tops the list of countries in the international PISA tests in Sciences and Mathematics, and two of the Lasallian schools are members of the elite group Hong Kong Grant Schools.

It was Thanksgiving Day back home. I was on a morning walk on the streets of Hong Kong, where it was another normal weekday. One of the many things that surprised me, including the beautiful and somewhat unsettling scaffolding structures constructed out of simple bamboo tied with plastic strings and covering gigantic high rise buildings, was the large number of young students, many walking alone, many with their friends, many with their parents, most wearing elegant school uniforms purposefully walking to school. 

School in Hong Kong’s culture is an enterprise, an endeavor that requires boldness and boundless energy. This commitment to excellence is rewarded with an outstanding education (Hong Kong is at the top of the international PISA tests in Science and Mathematics.) At the same time it levies the lives of families with insatiable demands of time and effort, not only for the children, but the parents as well. 

I was visiting with some of my research team colleagues at the Lasallian family of Schools in Hong Kong. There are eight Lasallian schools in Hong Kong. Two of them, La Salle College and St. Joseph’s College, are among the elite schools in the city. The Lasallian family has been in that part of the world since the beginning of the 20th century. Its history is marked, like the rest of the city, by long periods of prosperity and short but intense periods of crises and war. The classrooms where much learning has been engendered for generations were also the temporary barracks for invading and invaded armies, and generous makeshift hospitals. 

My research team and I collaborate with these schools in the area of Computational Thinking. Our work is done with public schools systems (mostly in the developing world), and with private school systems (mostly in the developed world). The governments of the developing world see Computational Thinking as their best chance to close the gap in their education systems, and the developed world sees Computational Thinking as the cornerstone of a technological society. 

England is the first major developed country to formally integrate into its public school system the subject of computation—the concept that supports the development of Computational Thinking. Beginning in Kindergarten and going through high school, children and young adults in England live the world of Computational Thinking every day. 

So, what is Computational Thinking? Why does it seem to attract such diverse attention, and what can it bring us that we don’t already have? 

The answers to these two questions are simple, and will be the subjects of the next paragraphs. But first a warning to the reader: The mind contains some cognitive traps that make some ideas difficult to see as they are, like an optical illusion. I will alert the alert reader when the mind may be tempted to throw a red herring. 

Heuristic Search

First an introduction to the workhorse of our so-called rational mind: the heuristic search. We think we are rational, but for the most part we are not. There are good evolutionary reasons why we could not survive with a reasoning mind—reasoning is slow, costly, and unreliable. The human species survived and prospered because it developed a decision-making process that was fast, inexpensive, and if not completely reliable, always favoring the safe response. 

S1: All professors are vain; some non-vain people don’t drink coffee; therefore some professors drink coffee. 

Sentence S1 is a simple argument with two premises and one conclusion. Computationally it is extremely simple, but many of us find it convoluted and taxing to our energy and patience. We certainly don’t yearn spending an afternoon playing that game. 

S2: The old banker and his wife bumped into his young mistress. 

Sentence S2 is computationally much more complex than S1, however our mind has computed effortlessly, automatically, immediately (and unavoidably) large sets of data, feelings, desires, reactions, and moods. 

In 2002, psychologist Daniel Kahneman was awarded the Nobel Prize in Economics for a simple, profound, and consequential idea: a) When we make decisions we think we reason, but we don’t, we instead use heuristics; and b) these heuristics are often biased and lead us to error. 

The reasons why we use heuristics were listed before. Why, however, we think we reason instead is perhaps a cognitive trap. To have an accurate model of the mind, our computational machine, is the first step in designing a productive educational system. 

Type-A problems 

Type-A problems are such an important set of problems that they deserve a name (perhaps a more descriptive name) and also a monument. A monument not because we admire them, rather a monument in the spirit of the Vietnam Wall Monument—something that reminds us of the past and makes us reflect about the future. 

During the K-12 years, Type-A problems are all that is done in Math and Sciences classes. We may have the impression that those long years cover a vast set of topics, but the reality is that it is only a variation on a single theme: how to solve Type-A problems. 

A Type-A problem consists of three phases: 

·Data: We are given a set of known data and unknown data, and we have to identify them. 

·Rules. There are rules that relate these data, and it is up to us to know them or to find them. 

·Result. we need to untangle the data attached to these rules and obtain the final result (this often requires algebra and almost always arithmetic). 

Type-A problems require for their resolution the activation of a set of cognitive processes known as System-2, which involves attention. System-2 tires soon, and it interferes, which means that no other task can be implemented at the same time. A surgeon would have to pause her open heart surgery, close her eyes, and deeply concentrate to correctly multiply 27 x 34 in her mind, if the life of her patient would depend on a correct answer (take a minute to experience multiplying these two numbers in your mind). That is a precarious system; a waste of the human capacities and training time. 

Computational Thinking

Each day we know more about how the mind works. What it can do and what it should not waste time trying to solve. Computational Thinking is the set of fundamental ideas, derived from cognitive sciences and computation, as well as epistemology, with two main goals: 1) Discover the uncharted world of cognitive processes that the human mind can sustain, provided the required tools; and 2) Develop a set of experiences that will allow students acquire these tools, engage in these cognitive processes, and produce rich and innovative results. 

Language and the Chinese Room

Ideas are supported in language, and complex ideas require languages that support complex ideas. Natural language—English, Spanish, etc.—is a miracle of evolution, and indeed allows for complex ideas, but as we saw earlier with sentences S1 and S2, they seem to work unevenly in different environments and soon find a limiting ceiling. Complex ideas require languages that support complex ideas. 

A language for representing complex ideas is based on a simple set of three mechanisms: 1) a set of primitives that we start with; 2) a means of combining these primitives; and 3) [this is the breakthrough that natural languages did not evolve to implement well] a means of abstraction by which we name the complex idea, and from now on it becomes a new primitive from which higher level ideas will be created. This the root of Emergence. 

The Chinese Room thought experiment deals with the concept of whether a computer can think and have cognitive states, as humans do. For the untrained mind this is a very difficult problem because we don’t know what the mind is, and we don’t know what a computer is. Some cognitive traps may supply some persuasive opinions, nonetheless unsupported. 

Philosophy and Psychology classes studying this problem start by first teaching students a language that supports the ideas of a Turing machine or a Von Neumann machine and how they operate (Engineering freshmen also learn this language, although with less philosophical objectives). This language provides the primitives to entertain in their minds models about their minds, and models about computers. 

It also allows students to entertain in their minds the fundamental idea of Emergence, how intelligence can emerge from non-intelligent devices (this is where the fundamental difficulty of the Chinese Room problem resides.) A NAND gate is an electronic device with two inputs. One output is always one, except when both inputs are one; at that time, the output turns to zero. Out of this simple, mindless, unintelligent device all the data and all computation in the world is born. 

Equipped with an appropriate language, the mystery that plagued philosophy classes for decades (and still does in some universities) has been downgraded to a simple problem that every student confidently solves. This is Computational Thinking. 

Curriculum Computational

Thinking provides the path to higher cognitive lives for our students and ourselves, but the path needs to be created. Students need to be immersed in the world where this language is spoken and where they can work with others and explore new ideas. It is similar to going to China for one year, getting integrated in Chinese society, learning their language and customs and culture and food and music by being with them and interacting with them. 

The world Computational Thinking is now being created by many research groups. It begins with the creation of curriculum—the activities that our children and young students will experience in their school life during their formative years. 

This work includes innumerable fields. My research team has been working for the last decade in areas such as Vector Differential Geometry (also known as the art of beautiful geometric), Cybernetics (the art and science of self-controlled organisms, such as butterflies and self-driving cars), Story Telling (the art of narrative self-expression), Discrete Calculus (the art of the moving universe), and others. 

A word on Discrete Calculus and the art of the moving universe: Long ago, Galileo concluded that a ball rolls down a ramp with a velocity that he described as the Law of the Odd Numbers (such as 1, 3, 5, 7, 9, …). That is probably one of the most beautiful laws in the universe. In fact, it rules the motion of all planets, stars, and galaxies in the universe. Later, Newton noticed that when subtracting the odd numbers the difference between them was constant (3-1=2, 5-3=2, 7-5=2) and that is the constant acceleration of the gravitational field. He also noticed that the cumulative sum was (1+3=4..+5=9,..7=16.. +9=25…) a set of square numbers: a parabola. 

This insight is not lost in a ten-year-old child, who soon creates a bouncing ball ruled by the Law of the Odd Numbers, which bounces like her basketball when she plays with her friends at recess. The formal name of the system that the child has created is a second order discrete differential equation system. This creation moves the ball as if it were in the presence of a real gravitational field. As the child plays with her creation, one day she notices that in the presence of an unexpected obstacle—a table, for example—the ball knows how high to bounce. Then she asks herself, “What is inside the Law of the Odd Numbers, which is the only thing the ball knows, that allows it know how high to bounce, regardless of the location of the obstacle?” The answer comes to her in less than one hour. This is Computational Thinking. 

These students live their school day in a cognitive world that is far from the unnecessarily fruitless and unnecessarily unpleasant world of convention. 

As I watch the children of Hong Kong briskly walking to their schools on a windy and bright morning, I wonder what places and experiences they will encounter today.

Dr. Olabe-Basogain is a professor of Electrical and Computer Engineering at Christian Brothers University

Posted by Josh Colfer at 1:32 PM

The Galleon is curated and managed by Christian Brothers University, a Memphis-based university founded in the Lasallian tradition (a sect within the Catholic faith). Part of our founding mission is to uphold respect for all persons-regardless of political, religious, or social beliefs. As an institution, we take no stand on political matters; to do so would undermine our commitment to intellectual inquiry and thoughtful response to events taking place in our World by members of the CBU community.

Share This: