Friday, May 31, 2013
Artificial Intelligence Explanation Even I Understand
I was honored when, earlier this week, my friend Adelson de Brito, a Professor of Physics in Brazil, decided to break down some very scientific information about Artificial Intelligence, in a way that even I, more of an artist than academic, can understand. Here, Adelson (who speaks Portuguese and many other languages) writes simply and beautifully about professor Andrew Ng’s research, AI, and algorithms — concepts I probably would have no hope of grasping had it not been for Adelson's fine craft as a teacher. Adelson’s students are blessed to have him. Thanks again Adelson! Here's the article: [Not long ago] I read [an article about] Dr. Andrew Ng. When he was a kid, he dreamed of building machines that could think like people, but when he got to college and came face-to-face with the Artificial Intelligence (or simply AI) research of the day, he gave up. Later, as a professor, he would actively discourage his students from pursuing the same dream. Later on, fortunately Professor Ng changed his mind back to his primary visions of AI. He reportedly claims his 180 degree navigation course correction took place when he ran into the “one algorithm” hypothesis, popularized by Jeff Hawkins, an AI entrepreneur who’d dabbled in neuroscience research. And the dream returned. Well, Andrew Ng (born 1976, Chinese: 吳恩達) is an associate professor in the Department of Computer Science and the Department of Electrical Engineering by courtesy at Stanford University, and he works as the Director of the Stanford Artificial Intelligence Lab. He also co-founded Coursera, an online education platform, with Daphne Koller. He researches primarily in artificial intelligence machine learning, and deep learning. His early work includes the Stanford Autonomous Helicopter project, which developed one of the most capable autonomous helicopters in the world, and the STAIR (STanford Artificial Intelligence Robot) project, which resulted in a Robot Operation System (ROS), a widely used open-source robotics software platform. Ng is also the author or co-author of over 100 published papers in machine learning, robotics and related fields, and some of his work in computer vision has been featured in a series of press releases and reviews. In 2008, he was named to the MIT Technology Review TR35 as one of the top 35 innovators in the world under the age of 35. In 2007, Ng was awarded a Sloan Fellowship. For his work in Artificial Intelligence, he is also a recipient of the Computers and Thought Award (Wikipedia, 2013). Well at this time and place I think of my friend Alicia Benjamin and will try to explain what is an algorithm. In mathematics and computer science, an algorithm is a step-by-step procedure, to be used for calculations as much as a cake recipe is worth for making cakes. Algorithms are used for calculation, data processing, and automated reasoning. It is an effective method expressed as a finite, concise list or set of well-defined instructions for calculating, for example, the value of a mathematical function. Starting from an initial state and an initial “input”, the instructions work as a combined set of actions designed to stimulate the whole process to proceed through a finite number of well-defined successive states, eventually producing an “output” and terminating at a final ending state. According to Ng, in the early days of AI, the prevailing opinion was that human intelligence derived from thousands of simple agents working in concert, what MIT’s Marvin Minsky called “The Society of Mind.” In his book Minsky brilliantly portrays the mind as a "society" of tiny components that are themselves mindless. To achieve AI, engineers believed, they would have to build and combine thousands of individual computing modules. One agent, or algorithm, would mimic language. Another would handle speech. And so on. In short, they believed the brains disposed of one algorithm to be used at a time and each one would deal specifically with the task that fits its particular nature. Well, to “reproduce” such a machine seemed an insurmountable feat. The good news as seen by Ng is the solidity of the concept introduced by Minsky aside with the concept of “one algorithm” hypothesis popularized by Jeff Hawkins. Deep Learning is a first step in this new direction. Basically, it involves building neural networks — networks that mimic the behavior of the human brain. Much like the brain, these multi-layered computer networks can gather information and react to it. They can build up an understanding of what objects look or sound like. Now it is time to raise a simple question: What are the primary ideas behind the quest of men after a society that offers more time to have fun and less time to spend with bothering tasks of a mechanical daily life? First, it was the Industrial Revolution that brought about the transition to new manufacturing processes. It occurred in the period from about 1760 to sometime between 1820 and 1840. This transition included going from hand production methods to machines, new chemical manufacturing and iron production processes, improved efficiency of water power, the increasing use of steam power and development of machine tools. The transition also included the change from wood and other bio-fuels to coal. The Industrial Revolution began in Great Britain and within a few decades had spread to Western Europe and the United States. Then the Information Age, that I will refer to by the acronym “IA” was advanced by a society marked by the capitalization on the computer microminiaturization advances, with a transition spanning from the advent of the personal computer in the late 1970s, to the Internet's reaching a critical mass in the early 1990s, and the adoption of such technology by the public in the two decades after 1990. Bringing about a fast evolution of technology in daily life, as well as of educational life style, the Information Age has allowed rapid global communications and networking to shape modern society. Next will be the Artificial Intelligence “AI” Revolution with its machines or software, and is also a branch of computer science that studies and develops intelligent machines and software. As we have seen by the topics we discussed above, the central problems (or goals) of AI research include reasoning, knowledge, planning, learning, communication, perception and the ability to move and manipulate objects. At the end of the day, industry is becoming more information-intensive and less labor and capital-intensive. This trend has important implications for the workforce; workers are becoming increasingly productive as the value of their labor decreases. However, there are also important implications for capitalism itself; not only is the value of labor decreased, the value of capital is also diminished. In the classical model, investments in human capital and financial capital are important predictors of the performance of a new venture. However, as demonstrated by Mark Zuckerberg and Facebook, it now seems possible for a group of relatively inexperienced people with limited capital to succeed on a large scale. To see Adelson's original blog post, along with references used for the piece, click here.