Name: デヴィッド・J・マラン - Visitas Thinks Big 2018 - ハーバード大学 (David J. Malan - Visitas Thinks Big 2018 - Harvard University)
Uploaded: 2021-01-14T10:33:11.000Z
Duration: 18 min 13 s
Description: VoiceTubeの動画で発音を聞きながら英語表現を覚えよう！学べる英語：

When I first got here as a freshman some years ago, I immediately grabbed, uh, gravitated toward things familiar to me, right?

過去形

Like I knew in high school that I was I liked history.

I really like my constitutional law class.

And so when I sat down with my proctor, my resident advisor, that first week, those first weeks of school and was thinking about what courses should I take and what should I concentrate in or major in?

I really stuck with those things that I already knew.

And so I signed up is ah, government concentrator ago of Major and I spent the next year and 1/2 of my life taking government classes and related fields and statistics and math and so forth.

And yet, when I was in high school, I still had this curiosity about other fields, and I rather postponed in my mind once I became a freshman.

All all studies some dramatic arts later on some, uh, some Latin later on, even then, some archaeology later on, a kind of kept thinking I'll get to my electives.

But first I'm gonna focus on just getting done.

And it was completely to the exclusion of computer science as well.

In fact, I was always a bit of a geek, but I never had taken any CS courses in high school.

I still remember kind of peering through the door of our computer science lab and wondering like what my you are.

So I thought geek, your friends were actually doing with their heads, hunker down and typing away on some keyboard.

And to me, it was just c++ or job over these these names that I had heard but didn't really know what they are.

I didn't really know what this other field Waas and I didn't take this class CS 50 my freshman year because I was frankly scared of it.

It had and still has to some extent, this reputation to beware because it's so unfamiliar to so many students myself included.

And it wasn't until my sophomore year that I finally got up the nerve to sign up for CS 50 after shopping it for one week.

But even then, the only reason I got my foot in the door was because the professor at the time let me sign up for the course pass fail because I think the crux of it for this field and a few others was that I feared failure not doing well in a world that was completely new to me.

I assume that every other kid in the class, it's surely been programming since here she was like, 12 years old.

And reality was that's actually not the case.

70% of the students and CS 50 were actually just like me and have never studied computer science before.

But eventually I figured out, and that first semester I figured out exactly what it was that this field was.

It wasn't about Java or C plus plus or even about programming.

It was really about problem solving and thinking more carefully, more methodically about problems and how you can divide and conquer problems in the real world, sometimes with the help of a computer.

But frankly, you don't even need the help of a computer to solve these kinds of problems.

You just need a an algorithmic way of thinking, amore, methodical way of thinking.

So while the bus these days have these kinds of devices in our pocket and when you actually use it for old school purposes, like making a phone call, you might open up your contacts and maybe, ah, little inefficiently scroll through the list of friends and family, starting with the A's all the way down to disease.

Or maybe a little more cleverly, you might do auto complete and start searching.

It's actually not all that different from our approach and yesteryear toe looking up people we know in this device, right?

There's a whole bunch of names alphabetically in this, and they have corresponding numbers.

And that's exactly the same thing that's going on in your iPhone or Android phone or the like.

So let's consider how the phone these days actually finds the friend or family member you want to call quite quickly in using.

This is the metaphor, really the derivative of what that wrote the origin of what that device is so I could open up this phone book.

I'm looking for someone like my friend Mike Smith and look down at the first page.

See that he's clearly not on it because Smith starts with s.

And so I just keep turning the pages one at a time, looking for Mike.

Now, obviously, this is pretty slow, and it's gonna take me forever to find him.

I had sort of I knew algorithms from when I was growing up.

So, uh, 2468 And this, too is gonna take a lot longer.

But of course, no one in this room is going to do that.

And our iPhones and android devices don't do that.

What would you do instinctively if you had a book that's even this big to find?

And, of course, I'm gonna end up in roughly, like the M section, give or take.

Could Smith again starts with s, but what do I know about Mike Smith?

Yeah, he's kind of over there to the right, because s comes after the middle of the phone book.

And so both physically and metaphorically can I divide and tear this problem in half, leaving just half of the phone?

But don't worry, I don't actually need this phone book anymore.

Now, of course, maybe I've gone a little too far this time and I mean the T section, not the SS.

But that's okay, because I can against terror.

The problem in half, leaving me now with maybe not 1000 pages or 500.

Maybe I'm already down to, like, 250 pages.

And just look at the progress I'm making through the phone book rather than one page at a time or two pages of the time, but apparently hundreds of pages at a time.

And if I can continue this divide and conquer, or if we call it by its fancier named binary search by meaning to, and we keep dividing in half, hence the two do we eventually end up hopefully with one page on which Mike Smith actually is now what does this mean to be more efficient?

It clearly is faster than just looking through every page or every other page off the phone book.

But if we just think about fundamentally what we just did, this isn't even relate to computers.

It just relates to thinking were carefully about a problem.

If we think about I saw that souvenir to go.

Uh, if we just think about what we just did well, let's just consider very generic, not mathematically oriented.

It all chart where this is like the size of the problem I care about.

And by size I mean, like numbers of pages in the phone book.

You can keep it the number of pages in the phone book and let's just call the vertical axis here time.

And what's the relationship among these three algorithms we just tried?

Well, the first algorithm was just one page at a time, and every page might take me one second or half a second or whatever.

So it's a very straight or linear relationship.

Meanwhile, if I do two pages at a time, it's still the same linear relationship.

I'm just going twice as fast, so the shape of that curve is the same shape, but it's lower because if I consider this many pages in the first algorithm, it took me this money steps.

But in the second algorithm takes me half a cz many steps because we're going to two at a time.

But what's the shape of that third and final and, frankly, much more intuitive algorithm that all of us have been taking for granted all these years?

Even if we've never studied computer science or programming or program, it turns out it's much doesn't get negative.

It turns out it's much more curved like this that happens to be called log arrhythmic.

But the point is that in order for this third algorithm to take Maur time, you really have to keep going and going and going and really see how far out it goes before it really makes a fundamental impact on the amount of time it's taking.

Put another way, if I double the number of pages in that phone book tomorrow, how many more steps would it take me to find that same person?

So if you double the number of pages, the amount of time required to find Mike Smith really just inches up ever so slightly.

It doesn't maintain this same relatively slow approach.

Now, we can actually do this in another way.

What's another algorithm we might have in a venue like this?

We'll kind of be nice to know how many people are here.

Old school into 123 45678 Okay, that's gonna take forever.

字幕リスト動画再生

デヴィッド・J・マラン - Visitas Thinks Big 2018 - ハーバード大学 (David J. Malan - Visitas Thinks Big 2018 - Harvard University)

awkward

eventually

approach

figure