Manifold Learning - part 1

Background: How many dimensions?

When we talk of dimensions in casual conversation, we often recall high school geometry. A point has zero dimensions, a line segment has one dimension (length), a rectangle has two dimensions (length & width), and a block has three dimensions (length, width & height).

We can also think of dimensions as a tuple, or a set of numbers, and this set of numbers describe something. For example, you can think of color in terms of Red, Green, and Blue components. We can say color has three dimensions (R, G and B). The same color can be represented with a different set of numbers; Cyan, Yellow, Magenta, and blacK. This time, color has four dimensions (C,Y,M, and K). If we are consistent with our set of numbers, we can describe many things. eHarmony supposedly has 29 dimensions to describe each person. It simply means they use 29 numbers to describe a person, whatever those numbers are supposed to measure.

Now analysing three dimensions is straightforward. We can turn it into graphs and plots, and it is easy to visualize. Four dimensions, a little harder but doable (look up color solid or color space sometime). But, 29 dimensions? How about 19,200 dimensions? We need help for those.

Why sign language?

I find sign language fascinating.

• It's a language unlike the any other languages I know.
• It uses a different set of body parts to communicate.
• Sign language linguistics is hard.
• Sign language recognition is hard(er).

And it surprises me.

For the longest time, I thought sign language used only the hands and arms. Only when I started reading about it and interacting with the Deaf, did it become clear that it uses the whole body. The face and body posture is important in discourse. That's the exciting part, we (I) don't know how important it is, yet. Sign language linguistics is in its infancy when compared to spoken and written linguistics.

FSL also opened my eyes to the plight of the Deaf. The most valuable lesson I have learned is this:

The most important thing about sign language is the Deaf person using it.

Video to Data

or How Do We Represent Video as an Input to Our Various Algorithms?

What is video anyway? To oversimplify, video is a series of images shown one after another. There are many display and storage formats; but, in essence, they are all just a set of images. How fast, how often the images are presented is measured in frames-per-second (fps). For example, television in usually shown at 30 fps. Each frame is one image or picture; so, television will show us 30 images* per second, one after another. You can think of video as time-series data.

Let's turn our attention to each frame (image) in the video. How many numbers do we need to represent an image? It depends on the size of the image. For illustration purposes, let's assume that our image is 160 pixels high and 120 pixels wide**. That means we have 19,200 pixels (160 x 120) to represent the image. If we have a color image/video, each pixel has color information -- what colors are present at that particular pixel of the image. Depending on how color was encoded, we could have 3 or 4 numbers to represent color. If we have a gray scale image, each pixel will only have the intensity information -- how dark or bright is that particular pixel. Thus, a gray scale image will need 19,200 numbers to represent it. If we treat video as time-series data, each data point will have 19,200 numbers associated with it. And that is exactly how the FSL recognition system we implemented treats video data. We can do this because all the images (frames) in a video has the same size.

* For TV, it's actually half-images per second. To keep the discussion simple, I'm ignoring that.

** It doesn't matter what storage format was used in the original video; at some point, it will have to be displayed on the screen, which has pixels.

Intelligence, Consciousness, and Life

Warning: This post is just me rambling.

Whenever I talk about my research sometimes the topic meanders to Artificial Intelligence. Specifically, two questions come up; (1) are these machines really thinking? And (2) will machines become as intelligent as people? It seems that people conflate three things in these discussions: Intelligence, Consciousness and Life.

Because us humans are Intelligent, Conscious and Alive, we often think of them as one property -- being Human. So when we say something is Intelligent, we automatically assume it also means that we are saying that something is Conscious and Alive. Thus, when we talk about Artificial Intelligence, sometimes this is confused with Artificial Consciousness and Artificial Life. And this is on top of the usual confusion with respect to Intelligence, Consciousness and Life.

Globe Tattoo on Fedora 12

How to get Globe Tattoo prepaid wireless working in Fedora 12
updated: added minor details, minor edits

It mostly just worked. By default, Network Manager and Modem Manager are installed. If not, install them first. I also installed the usb_modeswitch utility; but I didn't need it. The USB Modem is detected after about 10 secs.

• Create a new Mobile Broadband connection in Network Manager.
• Network Manager and Modem Manager detects Globe Tattoo as HUAWEI Technology HUAWEI Mobile
• Choose Philippines and Globe Telecom among the mobile broadband providers.
• Accept the default settings and Save.
• Edit the settings

After reading assorted forums and some trial and error, here are the settings in Network Manager that worked for me.

Using the Globe Tattoo Pre-paid SIM card.

• access number: *99#
• username: blank
• password: globe
• APN: http.globe.com.ph

Using my Globe Post-paid SIM card.
I inserted my Globe Post-paid card in the modem and it worked. I made sure that I had set my internet settings to TIME. (Send TIME to 1111 via text message)

• access number: *99#
• username: globe
• password: globe
• APN: internet.globe.com.ph

Other notes

• Markings on the USB Modem declare it as a Huawei E1552 HSDPA USB stick
• lsub shows it is a Huawei Technologies Co., Ltd. E620 USB Modem
• lsub shows the USB ID as 12d1:1001
• The USB modem also shows up as a storage device with drivers and utilities for Windows, Mac OS. I didn't try these out. I usually just unmount this.
• A fancy colored sticker covers the LED indicator. The sticker was easy to remove, just peel slowly.
• The LED flashes green when connecting to 2G network.
• The LED flashes blue when connecting to 3G+ networks
• If the LED is steady the modem has successfully connected to the network.
• Under the SIM slot, there is a microSD card slot. I have yet to try this.
• I have 64-bit Fedora 12 installed on a Lenovo Thinkpad SL400; that's a topic for another post.

...and now back to the Program

After dealing with technical and personal problems, we are back on the net. Apologies to my three readers!

Regular posting starts tomorrow. Thanks for dropping by!

Welcome Thinkpad SL400

I could not afford the T-series nor the X-series. I settled for the small business line of Lenovo, the SL-series These have just about the same features but at half (or less) the price. What are the trade-offs?

• The SL-series are heavier and bulkier. Instead of using lightweight metal parts, thick plastic parts are used.
• Lower end video card.
• No roll cage in SL-series. The T- and X-series have a metal roll cage around the LCD and body to protect the unit from abuse. You can feel the difference in the LCD; the SL-400 flexes more.
• The SL-400 has a high gloss finish on the lid and the screen. Ugh. I really don't like this.

There are probably others that I have missed. These are the ones that I have noticed.

On the plus side, I got this one without any OS installed. Which means I don't have to pay for software that I don't use (i.e. Windows). I installed Fedora 11 (64-bit) and so far so good. I'll reserve another post for the Fedora 11 experience.