Sonic Route 44 Lemon Berry Slush

It is 77 Fahrenheit. It is hot. It is sunny. It is Spring. You haven't seen the sun in months at this intensity. Your tongue is dry. Your throat is dry. You are driving. Where do you go? What do you do about your life problems?

YOU GO GET A ROUTE 44 LEMON BERRY SLUSH FROM THE SONIC ON NEIL ST, THAT'S WHAT YOU DO!!!!! DO IT!!!

Courier Breakfast

Ahhh, the smell of pancakes, coffee, omelettes, and bacon in the morning. And to have it in an old fashion place that seems to have some sense of history soaked into its seats. This is a winner here in terms of identity and character. Home-grown, no franchise, Champaign-Urbana place to eat. Breakfasts are great. I had the steak and omelette, there's other great stuff of course!

Check out the movie here [movie].

Check their website out too:
http://www.couriersilvercreek.com/couriercafe/

Texas Roadhouse, Champaign, IL

Standard Texas steak dinner here. They throw their peanut shells on the floor. Quite a busy place, very popular. Food is not bad. Not gourmet, but think of it as upper-class comfort meat food place. What was fun though, was the peanut shells and the noisy crowd. And they celebrate birthdays there by making the birthday person sit on a horse saddle, and yell out their name, and shout "Yee Ha!" They also do this quirky looking dance. Oh, and although they are Texas Roadhouse, they originated in Indiana. Check out the movie here [movie].

Crash!

It happened again! This is the second time I have witness a truck crashing into the train bridge across Springfield Ave. The first time was when we were walking along Springfield on the way home. And then, BAM! This truck's container hits the bridge because it was to low. That truck ended up stuck underneath because it was going fast enough so that the momentum got it through a little under the bridge. But then it was unable to move forward or backward anymore.

This one happened before we got there. But again, we were walking to work when we say this truck there. When will they ever learn? Is the signage of the bridge height inaccurate? Do truck drivers not know the height of their trucks? Do truck drivers not believe in the laws of physics? Are train bridges invisible to truck drivers? Is there a gravitational singularity in the Springfield train bridge that draws large bodies such as trucks to itself uncontrollably?

Stay tuned next time, when we find out the answers to...The Mystery of the Crashing Trucks

SPRING 2007!

It took a while lah. At first the weather turn hot, then it suddenly snow again. But now it look like it is here to stay lor. The flower all come out again. This shot is of a tree outside my apartment. The flower bloom white one. Quite the very impressive lah. More thing happening now.

Feedback about recent heated Intelligent Design debate

Here are some recent articles in CNN about the science and intelligent design debate. The first article [download] considers a scientists reasons for why he believes in God as a scientist. The second article [download] talks about the possibility that religion is simply a matter of evolutionary necessity.

Recently, this author has heard about some incidences which reflected a bias of non-religious scientists who think that religious people do not consider what the evidence is suggesting. This comment arose out of a scientific discussion on how our personal beliefs do affect the way we interpret objective data.

I think I will have to comment on this as I feel that the discussion has a very crucial point, and the remark made by non-religious scientists may not be a fair one. Firstly, I do not think that the discussion point was wrong. In fact, I agree with it. Beliefs most certainly bias our interpretation of the world, and of reality. In fact, that is fundamentally what belief is. It is attributing meaning and cause to what we see around us. In this sense, the idea that belief is a psychological state innate in human beings (since we do not seem to see animals that have beliefs) out of evolutionary necessity. One theory is that if we do not have something to believe in, human beings would not function, we would in fact go crazy. Thus, to resolve this state, we developed religion.

This might be true, it might not. Scientifically, we cannot prove that it does not exist. No science can prove that anything does not exist. I think scientists have to conceed to this. This author is a scientist too, and I think there is no reasonable argument to suggest that science can ever get past this problem of proving non-existence, at least with current empirical methods. Similarly, science cannot prove the non-existence of God.

But back to religion as an evolutionary outcome. If in fact this is true, it says nothing about whether God is real. Just because we think our ideas of God are a result of our neurons firing, which are a result of our DNA "directing" our neurons to wire in a certain way, that results in this "feeling" or state of thinking there is a God, these do not have bearing on whether God exists or not. If God exists, he exists whether I think he does or not, he exists whether my thoughts about his existence arise out of a reflection of truth or if they arise due to pure chance.

That being said, religious people, specifically Christians, do not necessarily disagree with the existence of evolution. In theory, God could implement evolution as his way things should work? Is it not also a theory that he did not? What I mean is, these are theories, not facts. And science has yet to show that it can prove that evolution = no God.

So, in sum, we all belief something. For the non-religious scientist, it is simply that they believe that all they can observe has no intelligent cause. For the religious scientist, we also need to look at the objective evidence that there is intelligent cause for everything we see. No one is spared this burden. And surely, which side we are on, will bias the way we see thing. The point is not to say the other side believes what they do because they are not objective. The point is to recognize how we ourselves are biased, and how others might have different biases, and consider the data together.

If God exists, then the data will show it, if indeed this method can show it. If he does not, then the data will show it too, if this method permits, however, we know that this method cannot show that something does not exist. So perhaps we should be thinking if there is ever any way to show that something does not exist?

Binding and Bandwidth

This is an idea about what might happen if different types of information were attended to. Consider this thought experiment:

There is an item A, and another item B. A and B both contain sub-features A1...An, B1...Bn. When we attend to A or B, we are in fact binding A1...An, and/or B1...Bn, to represent A, B.

Now, we have limited bandwidth. Which means, we can only process a limited amount of information at any one time. Consider for the moment that we can only process 4 bits of information. So, if we attend to A, we only process A1-A4, and if we process B, we process B1-B4. We could, by way of divided attention, process A1,A2,B1,B4. Assuming that there is minimal cost in having to dissociate between two different groupings of features (which is rarely the case, but lets just assume that this is possible for argument's sake). This also means, we do not process the other information about the other features that are present.

Now, consider another type of processing, or rather, another level. If in fact we process something called A-B. That is, we bring the binding function up from the item level of A and B, to a higher representation that binds both A-B. What would this result in terms of the amount of information we can process at a time?

This is now only 1 bit of information. We would have more capacity left over (3 bits) from our initial 4 bits. Furthermore, within the 1 bit, we might be able to reinstate the original A1-A4, and B1-B4 via past experience. However, we will suffer from interference in this case, since we did not explicitly process A1-A4 or B1-B4, but rather A-B. Thus, there should be a cost of attending to this higher level at the expense of the lower levels. Likewise, there is a cost of attending to the lower levels at the expense of the higher levels. This is also known in the literature as chunking.

Thus, in summary is that the level of binding should be inversely related to the bandwidth, or the amount of information we can process at any one time.

Unsupervised Learning

Here's an R code that implements unsupervised learning.

vdmulearning.R

Here's the code to display the hexagonal outputs you see in this page.

vdmhexplot.R

This is a specific instance of an unsupervised learning network used by Von Der Malsburg, hence VDM. He was interested in getting the network to exhibit similar behavior to what is observed about the human primary visual cortex. In humans, the primary visual neurons are organized in a columnar fashion according to their sensitivity and selectivity to visual line orientations. That is, each neuron in the primary visual cortex is maximally active for a specific orientation of lines that it receives visual signals from in the environmental space. Furthermore, these neurons are grouped together such that adjacent neurons are each sensitive to close orientations.



This R code implements the VDM network specifically using the following line orientation stimuli. The stimuli consist of 19 input units selectively made active (1 or 0) to give rise to "orientation". In fact, the input stimuli is realized in R as a matrix of 1s and 0s in the right positions.





At first, the network outputs a roughly clustered pattern of activity to a particular orientation (bottom left). But after several training iterations (about 100 cycles, which is quite fast!), it displays columnar organization (bottom right).













Interesting directions to pursue from this code are: object-level representation, color, moving stimuli, 3D representation, binding, repetition suppression.

Here's my paper which describes the model in greater detail [VDM.pdf].

Perceptron Neural Network: Backpropagation

Here's an R [http://www.r-project.org/] implementation of a backpropagation network.

trainnet_perceptron.R
testnet_perceptron.R

The network learns by propagating the input activity to the output layer, then comparing the resulting output with desired outputs. The difference is computed as an error which is backpropagated to the lower layers to effect a weight change that will reduce this error magnitude.

The network is then tested with original or distorted inputs. In general, this network can compute input-output mappings effectively (within network limits which are a function of the number of bits of information required to distinguish inputs, and the number of hidden layers and units). However, it is poor at generalization and distorted inputs compared to the Hopfield network.

Check out my paper that explains in greater detail [Backprop paper].
Also check out this website http://www.gregalo.com/neuralnets.html

Hopfield Neural Network

Here's an R [http://www.r-project.org/] implementation of the Hopfield, auto-associative network.

trainnet_hopfield.R
testnet_hopfield.R

Here's an brief on how it works. Every unit in the network is connected to every other unit (see weight matrix configuration in figure). Input patterns are used to trained the network using Hebbian learning. The network learns by additively changing its weights to reflect instances of unit co-activation. Unit dissimilarities and inactivations are ignored.

The network is then tested on original or distorted inputs, and it will robustly return one of the original trained inputs (within limits).

Check out my paper that explains in greater detail [Hopfield paper].
Also check out this website http://www.gregalo.com/neuralnets.html