Archive for the ‘Science’ Category

Michael Mann and Anthropogenic Climate Change

No Comments »

High level, but good…


Very cool video from NASA

No Comments »

I know, I’ve been away for a while. Lots going on, but I just saw this video and decided I needed to put a link to it on my blog.

Enjoy!


Wow! Freaky Fish

No Comments »

Not much to say beyond wow! Take a look

I saw a still photo of this on Pharyngula a couple of days ago, but, somehow, the transparent part didn’t sink in until I saw this video.


The Monty Hall Problem

No Comments »

I’ve seen this discussed several times. The question is whether or not you should switch your decision after you pick a door and Monty shows you another door that was the wrong door. It really seems that the status of a door you didn’t pick couldn’t possibly have an impact on whether the door you did pick was the right door, and in a sense that’s true. But the argument is that you should switch (that switching is the better decision two thirds of the time. Here’s a clip from Numb3rs that gives a better explanation than my lame attempt here.

Convinced?? Me neither. So it just happened to be the right decision when he demonstrated it. I follow the math but I wasn’t so convinced. Then I read a comment on a blog about someone simulating it before they were convinced. Well, being a programmer, I figured that would be the best way to settle it for myself. So I threw the following ruby program together.

$switch = 0
$stay = 0

def play
  win = rand(3)-1
  doors = [:goat, :goat, :goat]
  doors[win] = :car
  pick = doors.slice!(rand(3)-1)
  $stay += 1 if pick == :car
  $switch += 1 unless pick == :car
end

3000.times {play} 

puts "switch=#{$switch}  stay=#{$stay}"

I was actually convinced by the process of just writing the code (without running it). When putting the code together, I realized that my first pick would be right 1 out of three times, but on the other 2 times, since one wrong choice was eliminated, I would always be right when I switched (for the 2 out of three times I guessed wrong initially). You may have to think about that a bit, since I’m not always very good at wording things like this.

Better yet… I also wrote a javascript version of the same simulation.

The simulation graphs how many times you win by staying versus how many times you win by switching.

Click here to simulate 3000 rounds

Here’s the code ( Unfortunately, it’s a bit harder to see the logic in the code since there’s so much code around presenting the results dynamically):

function randomDoor() {
    return Math.floor(Math.random() * 3);
}
function oneRound() {
    doors = ['lose', 'lose', 'lose']
    doors[randomDoor()] = 'win'
    if (doors[[randomDoor()]] == 'win') {
        stay++;
    } else {
        change++;
    }
    div = document.getElementById('doorResults');
    if (change > 0)
    stayWidth = 300 * (stay / (stay + change));
    else
    stayWidth = 300;
    changeWidth = 300 - stayWidth;
    div.innerHTML = '<table cellspacing="0" width="300">' +
    '<tr height="20">' +
    '<td width="' + stayWidth +
    '" bgcolor="#66B">Stay ' + stay + '</td>' +
    '<td width="' + changeWidth +
    '" bgcolor="#fff"/></tr></table>' +
    '<table cellspacing="0" width="300">' +
    '<tr height="20">' +
    '<td width="' + changeWidth +
    '" bgcolor="#6b6">Change ' + change + '</td>' +
    '<td width="' + stayWidth +
    '" bgcolor="#fff"/></tr></table>';
    rounds++;
    if (rounds == 3000) {
        clearInterval(playInterval);
    }
}
function simulateDoors() {
    stay = 0;
    change = 0
    rounds = 0
    playInterval = window.setInterval(oneRound, 10);
}

Developing Intelligence : The Science of Mind-Reading: SVMs Extract Intentions from Neural Activity video

No Comments »

(note: I just noticed that this entry that I wrote on January 12th was never published, so… better late than never)
Developing Intelligence : The Science of Mind-Reading: SVMs Extract Intentions from Neural Activity video.

I recently posted a blog about scientist reading what a person was seeing by watching brain activity. That was amazing, but, as I recall, they had to calibrate to each individual. In this post referencing a 60 Minutes segment, we see a computer using fMRI to determine what a person was thinking about without calibration. Lot’s more interesting stuff beyond that.


Storm

No Comments »

I just listened to this on Pharyngula and enjoyed it very much, so I thought I’d post it here.

I guess many of us secretly wish we could spontaneously launch such a delightful tirade against the “woo” that seems to get so little scrutiny but leaves us incredulous, not only that it’s just been said, but even more so, that it doesn’t get questioned.

(I just updated the youtube reference. It appears the original was pulled by the poster. If you enjoy it you might want to check out more of Tim Minchin. He has some very funny stuff (warning, it’s not all exactly “family (or work) friendly)”.


Pharyngula: Soon, we’ll be reading your minds!

1 Comment »

I promise, not every post will be just a pointer to some other blog, but this is pretty incredible. ┬áResearches used MRI scans and were able to “read” what the subject was seeing by looking at the MRI.

Pharyngula: Soon, we’ll be reading your minds!.


Edit translation
Machine translation (Google):
Loading...
Copy to editor
or Cancel