At MBR04

Conferences are always a stressing occasion. You will meet some monotone personalities. You will question if these guys are lack of common sense for their daily life. Well, many of them do not need. You will think like them, extreme your mind to the end of its limitation. I feel I am a thinking machine. Food is just to give fuel to the machine.

MBR04 is a mixture of philosophers, cognitive scientists, physicists, computer scientists. This year’s topic is abduction, visualization and simulation. By all means, people discuss how to make a scientific discovery, how to build models and how to use model in scientific discovery. Many discussions open new insights to me.

Philosophy vs. Science

Philosophers seek the good statements from their thinking. Scientists seek formalized reasoning on a topic. I was surprised on some of the talks that end at an obvious point in the epistemological process. But it seems a philosophy thesis can be like that. The most important is to set up some arguments and statements. I am afraid that they swift the meanings of the concepts between the statement. So philosophy is not a science which can be formalized and reasoning in restrict logic. Philosophy is about everything, from more subjective point of view.

Logic vs. AI

The logic studied in philosophy is not the same as in AI. Philosophy studies the logic in people’s epistemological process. AI is from CS and studies how to use program to realize the logic process. So computational issues are studied in AI, but ignored in philosophy. But when an AI approach works, philosopher will come to discuss its philosophy meanings. Some examples: Turing machine, machine learning, etc.

The upper streams of AI

I found AI is an application of mathematics and philosophy. Both are like source of AI. If I need analysis ability, I need to look into mathematics and philosophy. To see how mathematician and philosophy think about my problem, I may get better ideas.

My slides show some interesting presentations in MBR04

Books to read

John WOODS and Dov Gabbay The Reach of Abduction: Insight and Trial, (with Dov M. Gabbay), volume 2 of A Practical Logic of Cognitive Systems, to appear with Elsevier/North-Holland.

Theo Kuipers, (2000) From instrumentalism to constructive realism, Synthese Library of Kluwer AP

Theo Kuipers, (2001) Structure in Science, Synthese Library of Kluwer, AP

Pre-print paper about monotonic segmentation

Without noticing my co-authors, I post our recent paper here. One thing is to show that I am working on some real research; the second thing is to attract attention from other researchers. I remember my post brought some discussion with Eanonn Keogh. Maybe that is the initial point of this paper. Abstraction of our paper:

Abstract. Monotonicity is a simple yet significant qualitative characteristic for system behaviors. It is a critical problem in qualitative modeling to robustly detect monotonic pieces from scattered data obtained by numerical simulation or experiments. We consider the problem of segmenting an array in up to K segments. We want segments to be as monotonic as possible and to alternate sign.We propose a quality metric for this problem, present a linear time optimal algorithm,and compare experimentally its performance to a linear time top-down algorithm. We show that our algorithm is considerably faster and more accurate using some simulation data.

Lenovo bought IBM PC business

Chinese manufacturer Lenovo bought IBM PC business for US$1.8 B. This makes Lenovo the third PC manufacturer after Dell and HP. Here is a good analysis article (recommended by Daniel).

It is a surprise but reasonable thing. For IBM, Lenovo is the best buyer to make the its PC business survive. For Lenovo, it got all the technology in this domain, and most important, the starting date to go to the world. People are expecting this date. Now it is today.

I do not know from when semi-conductor manufacturing becomes the strategic technology for China. But from the IST conference, after I attended the Chinese networking session, this trend is clear to me. China is seeking the technology for designing and manufacturing chips and high performance computer.

In contrast, India focuses on service-oriented industry, such as software engineering, produce engineering, account services, call-center services, health care services.

For the industrialized countries, the world competition pushes them to even higher end business. Innovation and, knowledge-oriented economy has to be their focus. The requirements to highly educated people will become more important. They also have to be an open society to attract people from the world to win the world competition.

My research tips to my students

It takes me time to speak to the students what-to-do and how-to. Now I put some how-to information on my web site. I should say before teaching a student how to do research, I need to teach them how to be a responsible person, i.e. taking the responsibility of the quality and schedule of the work assigned to them.

To be responsible is more important than being a genius. To select a partner for research project is the same. I need someone who can response my emails, who can finish what he/she promises me to do. So simple, but many can’t.

Identify bad students

Daniel and Bruce sent me a lot of their opinions about how to identify a bad student. They both tell me that there is no magic I can do to turn a bad student to a good one. Daniel said "forcefully organizing many meetings with the student often won’t help". I am happy to know that I do not need to be tough or strict to the students, because that would not help.

Here is Daniel’s metrics:
• cannot keep track of tasks assigned to him and be responsible for such tasks;
• lies to you about what has been done and what hasn’t been done;
• repeatedly ignores some of your phone calls or emails.

Here is Bruce’s metrics:

1. weak language skills (written and oral communication are weak with respect to technical topics -- even for many native speakers of English)
2. not open to research culture (student does not make an effort to talk to others about research and does not come to colloquia or group meetings)
3. unclear about objectives (supervisor does not portray the point of the research and how it fits in the big picture)
4. laziness, inability or unwillingness to focus (student does not come to work regularly, or dissipates research energy when there)
5. insufficient background (student does not have clear grasp of the essential elements of computer science and the required background for the research topic)
6. no research spark (student does not possess an innate ability to innovate, especially necessary for PhD.)

Google ads on my page

May you have noticed the google adds bar at the right side. I saw many grid computing ads. Obviously, google classifies my blog into grid computing. Well, a topic I did some investigate, but I am not going to dig into it.

Do I look like more e-businesser?

My online prototye: Online Learning System

I put an online learning system on my web site. This system is based on open source Moodle. It uses php and mySQL. Currently it has the functions to manage online courses, such as content management, student management, collaborative functions etc. We are developing it into a system to manage online experiments for online laboratory.