Understanding Linear Systems and Row

Operations

So, first equation is how we hold to 1. 2nd equation is how many shots we
wanna take or make, I guess, and so we're we're sort of able to see. The
second example we looked at was from finance, and this was something
objectives that we're trying to satisfy. There's 2 examples. But more
importantly, the reason I brought this slide out, is because I'm gonna I'm
gonna go over the general notation for a linear system in a moment.

It has no numbers in it, and so I just wanted to make sure we saw an

example with numbers before we got kinda abstract. So there it is. There's
the general form of a linear system of equations. The behavior, we call those
constants. The a, we call constants.

The x, we call variables. B, we call constants. And so what we're gonna do is

associate it with that new system. We're gonna take all of the coefficients.
We're gonna store them in a matrix.

We're gonna call them the coefficient matrix. We're gonna 2, absolutely. I
think I missed the second one. Correct? Negative one.

Okay. I do have a negative one. That would be on my 3rd variable. And then
what are we missing in here? Okay.

Thank you. 2nd row? Yes, who's back? 0, 2. 0, 2, 5.

We got this. This is not terribly difficult. And then my last row is yes. 1, 1, 1, 1.
Okay.

So we understand the notation. That's excellent. And then the constant

vector, what's that? Yes, sir. 2 negative 94.

Okay. We understand the notation. Amazing. Okay. So now what I'm gonna
do, I'm gonna get rid of this.

And what I'm doing is I'm sort of working backwards here. So I'm giving you
an augmented matrix, and I wanna see if you can give me a linear system or
the linear system corresponding to this augmented matrix. So this if this is
the augmented matrix of my linear system sigma of equations, what's the
first equation? Yes, sir? X1, x1.

X2. X1 plus x2 plus x3. And what do we have to make that equal to? 7. 7.

They go the second equation? Yes, sir. Yeah. It doesn't get any changes.
Okay.

Is that hard to solve? How would we solve this? Yes. Absolutely. Backward
substitution.

I already know x 3. I don't have to do any work to know the value of x 3. Just
one of the equations tells me x 3 has to be 4. I would take that. I would plug
it in here.

I would get 9. I know that x 2+9 is 12, which tells me x 2 has to be 3. Done. X
2 is 3, x 3 is 4, put those in here, and I figure out what x one is. Okay?

So systems that look like this, if we have an augmented matrix that has this
simple form where we've got a bunch of zeros kind of in the in the bottom
left, it really needs to be solved with what we call backward substitution. Is
this? This is an example of something that's in row echelon form. Okay. So
systems that are in row echelon form are easy to solve.

And so what we're gonna do is we're gonna say, how can I take the general
augmented matrix and put it in row echelon form so that I have something
that's easy to solve? That's what we're gonna do. We're gonna talk about
tricks that we can use, a procedure that we can use to take any augmented
matrix at all and turn it into a simple one like this that's easy to solve. That's
how we're gonna solve systems and equations. We're gonna take our
augmented matrix.

We're gonna play with it using what we call elementary row operations.
We're gonna get something that looks like this, and we didn't even have to
think about it. It was really easy for us to solve on this one. Okay. So that's
what's done.
So there are 3 things. What we're gonna be doing is we're gonna be
manipulating matrices to get them into what I call Roac's non form. I'll tell
you and give you a more formal definition of what that is in a moment.
There's 3 things that we are allowed to do. And I think it's gonna make more
sense for at least save a little bit of time if I actually can just keep a tablet.

And so the first thing I'm allowed to do, the first elementary row operation
that I'm allowed to do when I'm trying to put something in a row echelon
form is interchanging 2 rows. And so my chain notation here between these
two matrices, I'm taking the matrix on the left. I'm interchanging rows 1 and
3, r 1, r 3. I'm interchanging those 2 rows. And so what's gonna happen
when I apply an elementary row operation to this matrix?

Let's fill this matrix in. Elementary row operation on the left to interchange 2
rows. So if I interchange rows 1 and 3 in my matrix that I have up on the top,
what's gonna happen to yes. Yeah. And so what would my what would my
top row be?

They're gonna switch places. So the question was, what's gonna happen?
Well, if we're interchanging rows 1 and 3, the answer I like the way you put
that is, aren't they just gonna switch places? And absolutely. That's what it
means to interchange 2 rows.

They switch places. And so what's my top row gonna do? Yes. Yes. 1114.

And what's my autocorrect? 52. 12. Absolutely. Awesome.

So that's the first thing that we're gonna be allowed to do to matrices when
we're gonna get them into that nice echelon part is one of the things we're
allowed to do. One of the legal operations is interchange 2 rows. They just
switch places. Next thing that we can do, take a row and multiply every
number in that row by a constant by the same constant. I'm not gonna touch
any of the other rows with, with number 2 here.

I'm just gonna take any every single number in a particular row. I'm gonna
multiply them all by the same constant, and that's my second row operation.
So the notation that I have here in number 2 is I'm taking the matrix on the
left and I'm doing, a half row 1. So I'm taking everything in row 1, 1,
hammering it by a half, and then loading the result back into row 1. So
what's my top row gonna be if I brush everything in that row with a half?

Yes. Yep. So I'm gonna say 1, 0.5, negative 0.5, and there is a negative there.
We gotta make sure if there's one thing you understand how to do in this
course, it's this. Okay.

The last one is kind of the trickiest but also probably the most important.
What we're gonna do is we're gonna take a particular row and multiply it by
a constant. So in this notation here, I'm gonna go, okay, take row 1 and
multiply it by negative 1 half. And then what I'm gonna do is I'm gonna take
that and I'm gonna add it to row 3, and then we're gonna see what we get.
So I'm taking row 1 alone.

But what I'm doing is I'm taking the numbers in row 1, I'm multiplying them
by negative one half, and then I'm adding that to row 3, and then we're
gonna see what we get. So I'm gonna take row 1, multiply it by a constant,
look at all those numbers, add that to the bottom row, and then we see what
we get. K. So what happens what's my first answer here? I'm taking
everything in the top row here, multiplying it by negative a half, and then
adding it to the bottom row.

So let's just go over to the first column. If I go negative a half to what I have
at the top, add it to what I have at the bottom, what am I gonna get? 2. It's
negative to max. 0.

Absolute. So what I did is I said negative a half times 2. That's negative 1,

and then I add it to 1, and I get 0. K. Not everybody's with me, and that's
okay.

You you're all capable of understanding this, but we're not seeing it right
away. It's just something we gotta pay attention to later. Now what And then,
what's my next entry gonna be? Yes. It's a SORNER blue.

1.5. 1.5. I think that's right. Yeah. 1.5.

And then what's my last entry gonna be? Yes. Yes. Yes. Yes.

You have to wait 3. 3. Absolutely. Oh, well, we can so that's a good question.
Like, I only have a line with an augmented matrix, but you can do row
operations on any matrix.

So, if I was talking about an augmented matrix, then the idea would align
there. Okay. Okay. If you didn't understand, here's what we did. We did
negative 1 half, 2, negative 1, and 2.

So that's negative a half times 1, plus for o Okay. So you don't realize that's
another way of looking a lovely day. Okay? You have to do these. You have to
be able to do these to do these.

It's really easy to make mistakes, really easy to make mistakes, and so we
gotta be really, really careful when we're doing these raw operations. Okay.
Any questions? Yes. Would you do this with any, 2 rows right here?

Any questions? Yes. Would you do this with any, duos or a u r? Yes.
Absolutely.

Yep. I can do it with any duos. It doesn't just have to be the top of the clock
before we get any duos. Yes. That's a hard one.

Okay. We'll do that next. That's a really good question. Question was, how
do you choose the constants? I just picked arbitrary numbers I wanted to
illustrate the process.

I will give you like, I'll give you an idea of how to choose the constants in a
moment. Okay. There's some alternative notation for sort of documenting
the row operations that they Lily to put it on. Use whatever you want as long
as it's clear and consistent. You don't have to use exactly the notation I use.

Here are some things that my, my colleagues like to use. Whatever you do,
make sure it's clear and make sure it's consistent. Because if there is a
mistake in there, I wanna be able to say, okay. This person understood what
they were doing. They were trying to execute this row operation, and I see
that.

They just made a little typo like they forgot a half or something like that. So
do yourself a favor. Half or something like that. So do yourself a favor.
Document the row operations clearly and consistently.
And save yourself a headache down a little bit. Okay. So here's what we're
gonna do. Let's work through one one example. I wanna take this
augmented matrix, and I want it to run it through a series of row operations,
and I wanna see what we get.

Don't worry about where I'm at. Don't worry about how I chose the row
operations. We'll do that in the next slide deck. For now, we just need to
make sure we understand the mechanics. If I gave you row operations, make
sure you could apply them to a given matrix.

Okay? So we're gonna start off with our matrix. 1, 1 sorry. 1, 4, negative 5. 1,
2, 1, 1, 2, 7, 4, 5.

And what I'm gonna do is I'm gonna apply negative 4 r 1 plus r 2. We'll
probably gonna see what we get. Okay. So what we're doing here is we are
going to be which row is going to change? Yes.

R 2. R 2. So because I'm not changing the other ones and I don't confuse
myself, I gotta fill in the stuff that's not changing. I'm taking negative 4 over
1 and adding it to row 2. So row 2's gonna change.

And so what I'm gonna do just so I don't confuse myself is I'm gonna fill in
the stuff that doesn't change, and that is just gonna make me less likely to
make a mistake. So if I take negative 4 times the first row and I add it to the
second row, I gotta get my face. Like I said, it's really easy to make mistakes
here. And so I get the same thing. 2 of us are pretty old, human.

If you don't understand what I did there, that's okay. But you do you're all
capable of understanding it, but we do need to invest if you don't
understand what we did, we need to invest time at some point down the
road to figure out what I did. Yes? How come you don't know as far as one of
that? It's just sort of when I write something like this, it's always that rule
that's being effective.

Yeah. I just take it that way. We'll go here. We'll go here. Yes.

Great question. So the question is we we fiddled around with row 2, and it

changed from this to this. Really good question. If row 2 appears later in a
subsequent row operation, do I use this row 2 or do I use this row 2? The
answer is I use this one.

Once we apply a row operation, we never go back to a previous matrix. This

charging ahead. Okay? Great question. Like, that's that.

This is now row 2 forever on. There was another question here. Yeah? There
is. And what I'm trying to do is I wanted to wipe out everything.

I'll talk about this in a moment, but since you asked the question was why
did we choose this row operation? What I'm trying to do remember that nice
matrix we saw that I said was in echelon form? It had all zeros here. So what
I'm doing is I am selecting row operations to kill all the stuff below here and
turn it into 0. And then what I'm gonna do is I'm going to select the row
operation that fills that and turns it into 0.

That's what I'm doing. And I'll I'll make all of that very clear in a moment. I'll
give you an algorithm. But if we want to know what we're doing, we'll go for
it. Okay.

1st row operation was just number 3, add 1 row to another. Add a constant
with 1 row to another. Great question. The order matters. If we did the same
row operations in a different order, we get a different matrix.

Very good question. Go here. Yeah. I'll give you I have a slide where I explain
it. Yeah.

Great call. Good question. Okay. I think I'll take 1 or 2 more. The lovely
interaction is way more fun Yes.

The one that I that is being added to so how do we know the question was
with this row operation? How do they know it's row 2? The row that is being
added to is the one that's changed. So, like, if the line is just r two from 3
We'll get we'll we'll make a point for doing it slowly when we get there in
process. Okay.

Gotta know. Lucky interaction. Okay. Now what I'm gonna do, I'm gonna
apply a second row operation. Okay.

And, again, it's of the same form. Form 3, constant 1 row to another. So
we've got 5 r 1 plus r 3. So we're taking 5 row 1 and we're adding it to row 3.
So row 3 is gonna change.

When we did negative 4 r 1, add it to row 2, row 2 changed. Here we're

taking 5 r 1. We're adding it to r 3. R 3 comes after the plus, and so r 3 is the
one that's gonna change. Okay.

So we're taking 5 row 1, and we're adding it to row 3. So row 3 is gonna

change, and everything else stays the same. So let's go 1, 1, 1. This is the
stuff that's not changing. K?

Fill everything in that's not changing, and then deal with the stuff that is
changing once everything else is in order. And so if I take 5 times the first
row, this matrix, if I take 5 times this row, add it to this row, what do I get?
Yes. So it's just gonna like And for it. Yeah.

For it. Yeah. Same thing. Take the plates here. Are we all in a grace?

Notice And we're just gonna have to move over here. So I don't love doing it
this way. But we're gonna take this matrix, and we're gonna put it over here.
So the next row operation, don't worry about why I picked this one. Just
worry about that I picked it, is I'm gonna take row 2 and I'm gonna multiply it
by a constant.

So what rows are not gonna change? We're on our 3rd row operation. Now
what rows are not gonna change, and what rows are gonna change?
Operation. Now what rows are not gonna change and what rows are gonna
change?

Which row is not gonna rows are not gonna change. Yes. Row 2 is gonna
change. Row 1 and 3 will not. And so we're gonna go leave this stuff alone.

It's not changing. Row 2 is gonna change. Row 2 is gonna change. Row 123
will not. And so we're gonna go leave the self alone.

It's not changing. Okay. Now we're almost there. So what I'm gonna do is I'm
just gonna say negative 6 bar 2 plus star 3. We're gonna go through this one
real quicker.
Okay. So we're taking something. We're manipulating row 2. We're adding it
to row 3. So row 3 is the one that's gonna change.

And so I'm gonna leave row 3 along. 111670122512. 111-701-212-05-120 oh,

what's my last one? I go back with 1803. Okay?

It's okay if you don't see exactly what I'm doing, but I guarantee you're
capable of it if you just put that investment in time. So this is something you
need to understand. You are capable of understanding. It might just take you
a little bit longer than somebody else. Yes?

So can I always ask you to ask questions? I'm not sure if you're Let me
address that anymore. Great question. Oh, yes. Sorry.

No. I'm always adjusting the row after the plus. So you said for example, r
three, plus minus 3, r two. How is it? Good question.

I'm always gonna have the minus 6 or the 4 or whatever plus able to hear
them speak. Thank you. Yeah. Ultimately, we're trying to solve easy systems
here. That's what we're building for.

And this guy right here is an easy system to solve. The one we started with
is not, and so that's ultimately what we're doing. We're trying to put this into
I like how you put it, lowest form, and then it's gonna be easy to work with.
We don't exactly call it lowest form. Maybe we should.

We call it echelon form or produce pro echelon form, but a good way to
think about it is lowest form. Okay. I said one more, but I can't resist. We'll
go actually one more and then we move on. I've got slides for that.

Yeah. Great question. How do you know what operations? I'll tell you in a
moment. Okay.

So we started with the system. That was the system corresponding to the
augmented matrix. Kind of thing that we haven't said is, how do I know when
I take the system and I apply these elementary row operations? How do I go
with the solution that the system doesn't change? Right?

If I take a system, apply an elementary row operation to it, I get a new

system. So now I have 2 systems. The old one, the one that I manipulate. In
principle, because they're different systems, they can have different
solutions. Turns out, if you apply an elementary row operation, you do not
disturb the systems of equation.

So if I have one system here that was obtained via our elementary row
operation from that system here, the 2 systems will have exactly the same
solutions, and that's all. That means we can do elementary row operations
to take something or turn it into something easy, and we haven't messed
with the solutions. So we can use the solution in the earliest system in the
simplest system and know there's still solutions to the more labeled system.
Okay. And so this is kind of what I'm saying.

Right? If I start off with an complicated augmented matrix and I apply row
operations to it, every time I apply a row operation I have them create a new
system, turns out the solution to that new system is exactly the same as the
solutions to the one I started with. Don't just take my word for it right now. I
don't want to prove that. If you're interested in why it's true, we can talk
about it.

But for the purposes of this course, it's just just understand that this is the
case. We apply these role operations. We're we're transitioning from one
system to another. Turns out rows does this first one satisfy this condition?
Yes.

Does the second one satisfy this condition? Yeah, the first non zero thing I
saw was the first thing I saw, which was 1. How about this one? Oh, okay.
Cool.

So that's pretty easy to check. Number 2, if there are any rows that are
entirely zeros, they have to all be shoved into the bottom. I can have 3 rows
that are all zeros all the way across, but they have to be at the bottom. I
can't have a 0 row at the top, and then a couple non zero, and then a zero.
Anything that's all zeros has to be at the bottom.

No ifs, ends, or buts. And then the last thing I have a folks have a really hard
time kind of wording this. We mean by this is if I've got a row and it's 0, 0, 1,
and then a bunch of stuff, and then in my next row I have 0, and maybe I
have a 1, and then some stuff, okay, or maybe that's like a 4 or something
like that, you see how the first zero as we move down I think this is a good
way to put it. As we move down the matrix, the leading one, the first non
zero thing, the leading one that's what I mean when I say leading one is in
the first one that you see when you scan the row. As we move down the
matrix, the look at 4 matrices.

Have a look at the first one. Is this in row echelon 4? See any issues. Yes in
the back. Okay.

So there we have there's an example where I have 2 rows of zeros. I can

have as many as I want. They just have to be involved in plotting. Okay. So
once you get a little bit of practice, it's really easy to tell quickly whether or
not a matrix is in row echelon form.

Row echelon form is pretty easy to solve, but there's another level of
reduced row echelon form, and that's ultra easy to solve. So I see 2 hands.
We'll go here first and then there. Absolutely. Yeah.

But what I'm doing is taking a snippet of 2 rows. But if I have a matrix of 20, I
have to be able to do that every every step. K. One more shot. Wouldn't
matter.

Yeah. Good question. Doesn't matter how much that column, it's gonna be
zeros here, and it's gonna be zeros there. Row echelon form insists that
there's zeros below a leading one, reduced row echelon form. The only
additional thing is if I look above the leading one, I have to see zeros.

And if I see anything that's not a 0, I'm not. So we're gonna look at the same
matrices, and we're gonna say reduced. So we knew the first one, it was in
row echelon form In order to see if it's also in reduced row echelon form,
take all my leading ones, take it upwards, because all I see zeros. If that's
true, then I'll be in reduced row echelon form. Is our first one in reduced row
echelon form?

I guess. It is. I only have one leading one. Part c, we said, is in row echelon
form. Is it reduced?

Got it. No. Because if I look at any one if I look at this leading one, I look
above, I see a 2. That's not 0. Not cool.
If I go to this column right here, I go to the leading one. I look below, I see
zeros. I look up, I don't. Okay. How about the last one?

2 of our matrices are in reduced row echelon form. 1 of our matrices is in

row echelon form but not reduced row echelon form and then one of them is
in neither. Yes. Yeah. Great question.

If it be I flipped rows 2 and rows 3, I would have something that's in reduced

row echelon form, which is Yeah. That's a good question. If you can solve
some you can always turn something in row echelon form into reduced row
echelon form. And so it suffices to know how to do reduced row echelon
form. Good question.

So one more answer. I'll go 2 more and then we got it. Yes. This guy right
here? Yeah.

Yeah. It is a leading one. And so the condition is if you look up, you gotta see
zeros. Well, I can't look up, so I automatically satisfy that condition. So that
is a leading one.

It's just there's nothing above it, and so we automatically satisfy that
condition. Yes? How much work do we have to do to find a solution to this
system? 0. That's our unique solution.

It's done. Okay. So we Right? It's actually 0x1 plus 0x2 plus 0x3, which is
also known as 0, so my last equation is 0 equals 1. Can I find values of
x1x2x3 to make all 3 of these things true at the same time?

No. Because I would have to make 0 equal to 1 and we can't do that. And so
whenever you see a situation where you've got all the coefficients are 0 and
then the constant is non zero, you don't have a solution. Okay? And if we
had to add, like, 4 zeros then the bar and then the 7, in order to solve that
system, 0 would have to be equal to 7 and it's not.

So there's no solution. So when you look at this bottom row here in the
example b or part b, that is impossible to satisfy you by the what x one, x
two, x three are. You have no solutions, right, we created a formula for it, the
general solution. And so we had to introduce a free variable and then write
all the variables in terms of each c last time. So we had, like, x 1 is 2 1 c, x 2
is 64 Any guesses as to how I'm going to identify my tree area?

There's a general rule we'll use. Yes. 3 is equal to t. If I just wrote that down,
I get a 100 hands going up saying why'd you make x 3 the free variable, and
we see what it is about x 3 that's different from x 1 and x 2. Look at that
matrix.

What is it that's different? Yes. It's in both x 1 is not a free variable. X 2, I see
a second call, leading 1. So I'm not gonna have a leading variable.

X 3, there's no leading ones. And so I'm gonna anything whose column

doesn't And so now I have x one in terms of the free variable. Okay. So what
did I do? I had to pick a free variable, so I went to the columns that didn't
have leading ones, and I said those are my free variables.

So I went one here, so I'm gonna say 3rd column did have a leading one, so
I'm gonna make my 3rd variable a 3 variable. That's me making x three a 3
variable, and then go back to the first equation, sub this in, and get x one in
terms of, 3 variable. So anybody see what x two is gonna be? I'll write it
down. But do we see immediately what x two is gonna be in terms of the 3
variables?

2 t. Yep. And so what I'm gonna do is I'm gonna go x 2 minus 2x3 equals 0.
This tells me that x 2 minus 2 t is equal to 0, and this tells me that x 2 is 2 t.
Okay.

So I have a general solution. Yes. Essentially, in any field that's a good

question. Anytime you have a column so if I have a situation where
everything is a leading one, every column has a leading one in it, I have a
leading solution. If I have a situation where there is a column that does not
have a leading one, I know I have it in the middle.

That's sort of the rule of thumb. Okay. Got to move on. Let's go to part d.
And Absolutely.

And then my second equation is just gonna be x3+5x4 is equal to negative.

Okay. And so again, I've got 2 very 2 columns of leading ones. So I'm gonna
have 2 variables that are not free variables. I've got 2 columns that don't
contain leading ones, and so I'm gonna have 2 free variables.

So here we have again, because I have variables that are appearing in

multiple equations, you know, because I have columns that don't
consistently don't don't have a lean one in them, I'm going to have an infinite
number of solutions, and I'm and I'm looking for a general solution. So what
do we do when we have 2 free variables? How do we know we have 2 free
variables? There's 2 columns that don't have any ones. So I have 2 free
variables.

That's how I know. So what are my free variables? So what I'm gonna do is
I'm just gonna say let what do I wanna do? Let's use s and t. Let's use s for
our first for barriers and t for our second.

What's my first day? Yes. That's 4. Absolutely. So how do we know this?

Well, column 2 does not have leading 1, so it's a free variable. You call it
variable 2. Column 4 does not have leading 1, so variable 4 is gonna be a
free variable. That's how we look for where the leading ones are not, and
those are our 3 variables. Okay.

And so what I'm gonna do is I'm gonna leave it to you. I'm just gonna write
down what the for the generous solutions, why did we, choose x 2 and x 4?
Can anybody answer that? The question was, why do we choose x2 and x4
as the free variance? Yes.

They're not they don't have leading ones and all these problems. Yes. Yes.
Yeah. There's no concept of a free variable if you don't have an infinite
number of solutions.

Is that helpful? Okay. Yes. If you can see the one at 5, if you can see the top
faces, you know, the first one there, whether that helps. So and then we've
got a yes.

Solution, do we have to write it as a set or are we also able to write it, like, or
do we have to write it like a set? You can write it like as I did before where I
said, like, this comma that comma that. Yeah. We can do each one. Yeah.

Yeah. Can't hear the question. Or you don't you don't need to speak up.
Everybody else needs to come down. We'll wait until that happens.

Do multiple once a month, can I do it? The answer is yes. At the end of the
day, if you get the right answer and it looks like you knew what you were
doing, it'll be fine. The only problem comes with it, if you make a small
mistake, if you're doing 2 row operations at once and you just make a small
mistake, it might be hard for me to see what you're trying to do. And so
that's simply risky.

Okay? Alright. I gotta move on. I've got 15 minutes and still some stuff to talk
about. So here we go.

Identifying 3 variables. Probably should have had this slide up before. But
this answers all this question. If there is a leading one, then we're not a free
variable. If there is not a leading one, then we are a free variable.

Okay. And so there's 2 types of variables. There's 3 variables. There are

variables. There's 3 variables.

There are 3 variables. That's just the terminology I'd love for you. I might
not. In any event, this is how we find our 3 variables. If I have a 3 variable, I
have an infinite number of solutions, and we know how to write it down.

Last piece of the puzzle. So what have we talked about? We know how to
execute row operations. We know how to identify things that are in reduced
row echelon form. We know how to solve systems that are in reduced row
echelon form.

The only thing we don't know how to do is take something that's not in
reduced row echelon form and turn it into something that is in reduced row
echelon form. That's the last piece of the puzzle. The first thing that we're
gonna do is take something that's not in reduced row echelon form. That's
the last piece of the puzzle. That is in reduced pro echelon form.

That's the last piece of the puzzle, and we're gonna use what we call dose
Jordan elimination. I'm gonna go through this here. It's very straightforward.
This is not something that students and and in my experience, if you do
struggle with it, it's not because you're not capable of understanding it. It's
probably you just didn't invest the time figuring out how to do it.
There's all sorts of resources online. I'm gonna unfortunately have to rush
through a little bit, about 15 minutes. It's not quite as much time as I wanted.
But ex Jordan's elimination is something that you need to be able to do. And
if you can't follow what I'm doing here, I think you'll be able to.

There's all sorts of resources out there online, and maybe one of them
speaks to you in a way that doesn't explore MOS. I'm not even gonna explain
this algorithm. I'm gonna show you an example and just kind of talk about
what I do. There it is written down. Much easier if we illustrate it through a
series of So here we go.

So what I always want, I always want that leading one of them stopped
working. Don't win. What has to happen below a leading one? 0. They all
have to be zeros.

So what do I what's my first move? Another ball, a shoe, and then click.
What do I what's my first move? And I need to choose operations that are
gonna do what? Is that a hand no tentative?

I just wanna try and get see if I can find somebody who's an ant has an ant. I
love the participation. I just wanna find a new voice. What? So there's a 3 r
one, you're this close.

If I go 3 r one plus r two, I got 3 plus 3 and 5. Minus. Absolutely. So what I

wanna do, I gotta turn that into 0. I gotta turn that 0.

So that's gonna be my first two moves. Right? Start with the leading one. We
wanna clear zeros. Right?

We want zeros. K? So we start here. We clear everything out. So in order to

get this to go away, I gotta go negative 3 times this plus that.

If I'm at negative 2 times this plus that, I get 1. That's not what I want. My
first row operation needs to turn this into 0. Negative 3 times this plus that
gives me 0. So we go negative 3 r 1, add it to r 2, and what am I get again?

Let's fill in my row. Now I'm not Yes. It's exactly what it is. K. I made Negative
2roneplusr.
Absolutely. Negative 2rone. Let's go up here. So in your notes, like, I would
be continuing down there, but I just gotta place some space over there. So
what's the next thing to do?

I looked at my first leading one, and I cleared everything up and turned it

into 0. Now I'm gonna go find the next leading one. So what's my next move
gonna be? You're gonna take 3 r 2 and add 2 r 3. 3 r 2.

Okay. I'm gonna slow it down a little bit. I think you're probably moving 2
steps where I just wanna move 1. Yes. 1 half r two.

1 half r two. So I'm gonna slow it down a little bit. I have a leading one. That
should be I want a leading one here. That's not a 2.

I just need to turn that into a 1, and then I have my next leading one. And so I
started with a leading one. I cleared everything out. Now I need to go find
my next leading one. That's a leading number, but it's not a one.

So I gotta turn it into a 1, and that's where the multiply by row by a constant.
That row operation where we multiply row by a constant, we do that to
create leading ones. Okay. So we are gonna go one half r 2. And now I have
my next leading one.

Okay. Yes. The, negative 3 r 2 plus, plus r 3. Absolutely. You got to take care
of 1 of them.

Negative 3 r 2 plus r 3. That's gonna give me 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1,

2, 0, 0, 1, 3. Okay. So what I'm gonna do is I'm just gonna say we are running
out of time, and so I'm just gonna sort it. Beautiful leading 1, 0s above below.

Beautiful leading 1, 0s above below. Leading 1. I need 2 more row operations.

We'll just I'll write down what I did and then give you the final digits. Yeah.

Well, we're always gonna do, like this is always gonna be a plus. So I'm just
gonna reverse what he said and say minus 2 r 3 plus r 2. And then what's my
last move? Yes. R 1.

Yeah. And so my last moves are after minus 2 r 3 plus r1, we get this. Secret
matrix and put it in reduced pro echelon form. It's another thing to be able to
take something in reduced pro echelon form and solve the associated
system, and so I'm gonna see both here. Okay.

And so we could say something like the inclusion is x1 equals 5, x2 equals

negative 2, x3 equals 3. Or we could say something like x1x2 x3 is equal to 5
minus 2. K. So what I'm gonna do is we are as soon as I say we're out of
time, that I use to put b and c into reduced row echelon form, and then I'm
gonna put the reduced row echelon form in. It's up to you to make sure that
you can derive that sequence of row operations yourself, but use what I do
as a check on your work.

And I guess what I'm saying is, do we see in your own? Now put the answers
up online. Everybody in here is making this investment in time. Everybody
has the right to hear the important things. Okay.

So I will put the I expect you to do b and c. That's my expectation. I will put
the answers up. I'll post them on my learning space. I also forgot to include a
couple of slides about matrix rank.

I will include those under today's notes or the notes, and I'll expect you to
read those.