Finding a certain entry in a matrix
Clash Royale CLAN TAG#URR8PPP
up vote
5
down vote
favorite
Can we find the entry $s_23$ of $S=H^3$ where $H=pmatrix2&-1&0\3&1&2\-1&1&1
$ without finding $S$. I know that $s_23$ is given by multiplying the second row of $H^2$ and the third coloumn of $H$ but it is dull. Thank you for any hints!
linear-algebra matrices
add a comment |Â
up vote
5
down vote
favorite
Can we find the entry $s_23$ of $S=H^3$ where $H=pmatrix2&-1&0\3&1&2\-1&1&1
$ without finding $S$. I know that $s_23$ is given by multiplying the second row of $H^2$ and the third coloumn of $H$ but it is dull. Thank you for any hints!
linear-algebra matrices
add a comment |Â
up vote
5
down vote
favorite
up vote
5
down vote
favorite
Can we find the entry $s_23$ of $S=H^3$ where $H=pmatrix2&-1&0\3&1&2\-1&1&1
$ without finding $S$. I know that $s_23$ is given by multiplying the second row of $H^2$ and the third coloumn of $H$ but it is dull. Thank you for any hints!
linear-algebra matrices
Can we find the entry $s_23$ of $S=H^3$ where $H=pmatrix2&-1&0\3&1&2\-1&1&1
$ without finding $S$. I know that $s_23$ is given by multiplying the second row of $H^2$ and the third coloumn of $H$ but it is dull. Thank you for any hints!
linear-algebra matrices
linear-algebra matrices
asked Sep 6 at 6:48
B.B.
1467
1467
add a comment |Â
add a comment |Â
2 Answers
2
active
oldest
votes
up vote
9
down vote
accepted
If you don't want to calculate $H^3$ or $H^2$, you can take the vector $v=beginpmatrix0&0&1endpmatrix^T$ and calculate $w=H cdot H cdot H cdot v = H cdot H cdot beginpmatrix0&2&1endpmatrix^T$. The entry $s_23$ will be the second coordinate of $w$.
You need 18 multiplications and 12 additions to calculate this. Calculating $H^2$ would need 27 multiplications and 18 additions. I don't know whether this is still too 'dull' for you ;)
This works since $s_23$ is the $e_2$-component of the image of $e_3$ under $S$.
Edit: Even less work is needed by a comment of celtschk. We may extract the $e_2$-component of $H H beginpmatrix0&2&1endpmatrix^T$ by multiplying with $beginpmatrix0&1&0endpmatrix$ from the left. Therefore, we can calculate $$s_23 = beginpmatrix0&1&0endpmatrix HH beginpmatrix0&2&1endpmatrix^T = beginpmatrix3&1&1endpmatrix H beginpmatrix0\2\1endpmatrix$$ with 12 multiplications and 8 additions.
2
Even less work is needed: $$pmatrix3&1&2cdot Hcdotpmatrix0\2\1.$$ 12 multiplications and 8 additions.
â celtschk
Sep 6 at 15:40
@celtschk You are totally right. I'll edit this
â Babelfish
Sep 6 at 15:58
How are you counting to get the $12$ multiplications and $8$ additions?
â Jose Brox
Sep 7 at 11:00
Multiplying $H$ with the left vector yields $3times 3$ multiplications. We have to add two of such products to get the entries of the result, which yields $3times 2 $ additions. Multiplying the two remaining vectors yields 3 multiplications and two additional additions. Of course, if we don't count the multiplications with $0$ at the beginning, it's only $9$ multiplications and $6$ additions.
â Babelfish
Sep 7 at 11:13
add a comment |Â
up vote
7
down vote
The characteristic polynomial of $H$ is $$X^3-4X^2+6X-3.$$
Therefore $$H^3=4H^2-6H+3I=H(4H-6I)+3I.$$
The element $(H^3)_23$ is then the product of the second row of $H$ times the third column of $4H-6I$ (plus nothing since $3I$ has element $(2,3)$ equal to $0$).
We need $18$ products and $12$ additions to get the polynomial, plus the expanding of terms. After that we need 1 addition (from the -6I), 3 multiplications (from the 4H), plus 3 multiplications and 2 additions from the second row-third column product.
But hey, this way is not dull!
1
Definitely not dull. I like creative usage of Cayley Hamilton! +1
â Babelfish
Sep 6 at 8:16
Is $A=H$? If so, you may want to consistently use $H$ throughout.
â Meni Rosenfeld
Sep 6 at 12:52
@MeniRosenfeld Yes, of course. Gonna correct it. Thanks!
â Jose Brox
Sep 6 at 12:58
add a comment |Â
2 Answers
2
active
oldest
votes
2 Answers
2
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
9
down vote
accepted
If you don't want to calculate $H^3$ or $H^2$, you can take the vector $v=beginpmatrix0&0&1endpmatrix^T$ and calculate $w=H cdot H cdot H cdot v = H cdot H cdot beginpmatrix0&2&1endpmatrix^T$. The entry $s_23$ will be the second coordinate of $w$.
You need 18 multiplications and 12 additions to calculate this. Calculating $H^2$ would need 27 multiplications and 18 additions. I don't know whether this is still too 'dull' for you ;)
This works since $s_23$ is the $e_2$-component of the image of $e_3$ under $S$.
Edit: Even less work is needed by a comment of celtschk. We may extract the $e_2$-component of $H H beginpmatrix0&2&1endpmatrix^T$ by multiplying with $beginpmatrix0&1&0endpmatrix$ from the left. Therefore, we can calculate $$s_23 = beginpmatrix0&1&0endpmatrix HH beginpmatrix0&2&1endpmatrix^T = beginpmatrix3&1&1endpmatrix H beginpmatrix0\2\1endpmatrix$$ with 12 multiplications and 8 additions.
2
Even less work is needed: $$pmatrix3&1&2cdot Hcdotpmatrix0\2\1.$$ 12 multiplications and 8 additions.
â celtschk
Sep 6 at 15:40
@celtschk You are totally right. I'll edit this
â Babelfish
Sep 6 at 15:58
How are you counting to get the $12$ multiplications and $8$ additions?
â Jose Brox
Sep 7 at 11:00
Multiplying $H$ with the left vector yields $3times 3$ multiplications. We have to add two of such products to get the entries of the result, which yields $3times 2 $ additions. Multiplying the two remaining vectors yields 3 multiplications and two additional additions. Of course, if we don't count the multiplications with $0$ at the beginning, it's only $9$ multiplications and $6$ additions.
â Babelfish
Sep 7 at 11:13
add a comment |Â
up vote
9
down vote
accepted
If you don't want to calculate $H^3$ or $H^2$, you can take the vector $v=beginpmatrix0&0&1endpmatrix^T$ and calculate $w=H cdot H cdot H cdot v = H cdot H cdot beginpmatrix0&2&1endpmatrix^T$. The entry $s_23$ will be the second coordinate of $w$.
You need 18 multiplications and 12 additions to calculate this. Calculating $H^2$ would need 27 multiplications and 18 additions. I don't know whether this is still too 'dull' for you ;)
This works since $s_23$ is the $e_2$-component of the image of $e_3$ under $S$.
Edit: Even less work is needed by a comment of celtschk. We may extract the $e_2$-component of $H H beginpmatrix0&2&1endpmatrix^T$ by multiplying with $beginpmatrix0&1&0endpmatrix$ from the left. Therefore, we can calculate $$s_23 = beginpmatrix0&1&0endpmatrix HH beginpmatrix0&2&1endpmatrix^T = beginpmatrix3&1&1endpmatrix H beginpmatrix0\2\1endpmatrix$$ with 12 multiplications and 8 additions.
2
Even less work is needed: $$pmatrix3&1&2cdot Hcdotpmatrix0\2\1.$$ 12 multiplications and 8 additions.
â celtschk
Sep 6 at 15:40
@celtschk You are totally right. I'll edit this
â Babelfish
Sep 6 at 15:58
How are you counting to get the $12$ multiplications and $8$ additions?
â Jose Brox
Sep 7 at 11:00
Multiplying $H$ with the left vector yields $3times 3$ multiplications. We have to add two of such products to get the entries of the result, which yields $3times 2 $ additions. Multiplying the two remaining vectors yields 3 multiplications and two additional additions. Of course, if we don't count the multiplications with $0$ at the beginning, it's only $9$ multiplications and $6$ additions.
â Babelfish
Sep 7 at 11:13
add a comment |Â
up vote
9
down vote
accepted
up vote
9
down vote
accepted
If you don't want to calculate $H^3$ or $H^2$, you can take the vector $v=beginpmatrix0&0&1endpmatrix^T$ and calculate $w=H cdot H cdot H cdot v = H cdot H cdot beginpmatrix0&2&1endpmatrix^T$. The entry $s_23$ will be the second coordinate of $w$.
You need 18 multiplications and 12 additions to calculate this. Calculating $H^2$ would need 27 multiplications and 18 additions. I don't know whether this is still too 'dull' for you ;)
This works since $s_23$ is the $e_2$-component of the image of $e_3$ under $S$.
Edit: Even less work is needed by a comment of celtschk. We may extract the $e_2$-component of $H H beginpmatrix0&2&1endpmatrix^T$ by multiplying with $beginpmatrix0&1&0endpmatrix$ from the left. Therefore, we can calculate $$s_23 = beginpmatrix0&1&0endpmatrix HH beginpmatrix0&2&1endpmatrix^T = beginpmatrix3&1&1endpmatrix H beginpmatrix0\2\1endpmatrix$$ with 12 multiplications and 8 additions.
If you don't want to calculate $H^3$ or $H^2$, you can take the vector $v=beginpmatrix0&0&1endpmatrix^T$ and calculate $w=H cdot H cdot H cdot v = H cdot H cdot beginpmatrix0&2&1endpmatrix^T$. The entry $s_23$ will be the second coordinate of $w$.
You need 18 multiplications and 12 additions to calculate this. Calculating $H^2$ would need 27 multiplications and 18 additions. I don't know whether this is still too 'dull' for you ;)
This works since $s_23$ is the $e_2$-component of the image of $e_3$ under $S$.
Edit: Even less work is needed by a comment of celtschk. We may extract the $e_2$-component of $H H beginpmatrix0&2&1endpmatrix^T$ by multiplying with $beginpmatrix0&1&0endpmatrix$ from the left. Therefore, we can calculate $$s_23 = beginpmatrix0&1&0endpmatrix HH beginpmatrix0&2&1endpmatrix^T = beginpmatrix3&1&1endpmatrix H beginpmatrix0\2\1endpmatrix$$ with 12 multiplications and 8 additions.
edited Sep 6 at 16:02
answered Sep 6 at 7:18
Babelfish
1,004115
1,004115
2
Even less work is needed: $$pmatrix3&1&2cdot Hcdotpmatrix0\2\1.$$ 12 multiplications and 8 additions.
â celtschk
Sep 6 at 15:40
@celtschk You are totally right. I'll edit this
â Babelfish
Sep 6 at 15:58
How are you counting to get the $12$ multiplications and $8$ additions?
â Jose Brox
Sep 7 at 11:00
Multiplying $H$ with the left vector yields $3times 3$ multiplications. We have to add two of such products to get the entries of the result, which yields $3times 2 $ additions. Multiplying the two remaining vectors yields 3 multiplications and two additional additions. Of course, if we don't count the multiplications with $0$ at the beginning, it's only $9$ multiplications and $6$ additions.
â Babelfish
Sep 7 at 11:13
add a comment |Â
2
Even less work is needed: $$pmatrix3&1&2cdot Hcdotpmatrix0\2\1.$$ 12 multiplications and 8 additions.
â celtschk
Sep 6 at 15:40
@celtschk You are totally right. I'll edit this
â Babelfish
Sep 6 at 15:58
How are you counting to get the $12$ multiplications and $8$ additions?
â Jose Brox
Sep 7 at 11:00
Multiplying $H$ with the left vector yields $3times 3$ multiplications. We have to add two of such products to get the entries of the result, which yields $3times 2 $ additions. Multiplying the two remaining vectors yields 3 multiplications and two additional additions. Of course, if we don't count the multiplications with $0$ at the beginning, it's only $9$ multiplications and $6$ additions.
â Babelfish
Sep 7 at 11:13
2
2
Even less work is needed: $$pmatrix3&1&2cdot Hcdotpmatrix0\2\1.$$ 12 multiplications and 8 additions.
â celtschk
Sep 6 at 15:40
Even less work is needed: $$pmatrix3&1&2cdot Hcdotpmatrix0\2\1.$$ 12 multiplications and 8 additions.
â celtschk
Sep 6 at 15:40
@celtschk You are totally right. I'll edit this
â Babelfish
Sep 6 at 15:58
@celtschk You are totally right. I'll edit this
â Babelfish
Sep 6 at 15:58
How are you counting to get the $12$ multiplications and $8$ additions?
â Jose Brox
Sep 7 at 11:00
How are you counting to get the $12$ multiplications and $8$ additions?
â Jose Brox
Sep 7 at 11:00
Multiplying $H$ with the left vector yields $3times 3$ multiplications. We have to add two of such products to get the entries of the result, which yields $3times 2 $ additions. Multiplying the two remaining vectors yields 3 multiplications and two additional additions. Of course, if we don't count the multiplications with $0$ at the beginning, it's only $9$ multiplications and $6$ additions.
â Babelfish
Sep 7 at 11:13
Multiplying $H$ with the left vector yields $3times 3$ multiplications. We have to add two of such products to get the entries of the result, which yields $3times 2 $ additions. Multiplying the two remaining vectors yields 3 multiplications and two additional additions. Of course, if we don't count the multiplications with $0$ at the beginning, it's only $9$ multiplications and $6$ additions.
â Babelfish
Sep 7 at 11:13
add a comment |Â
up vote
7
down vote
The characteristic polynomial of $H$ is $$X^3-4X^2+6X-3.$$
Therefore $$H^3=4H^2-6H+3I=H(4H-6I)+3I.$$
The element $(H^3)_23$ is then the product of the second row of $H$ times the third column of $4H-6I$ (plus nothing since $3I$ has element $(2,3)$ equal to $0$).
We need $18$ products and $12$ additions to get the polynomial, plus the expanding of terms. After that we need 1 addition (from the -6I), 3 multiplications (from the 4H), plus 3 multiplications and 2 additions from the second row-third column product.
But hey, this way is not dull!
1
Definitely not dull. I like creative usage of Cayley Hamilton! +1
â Babelfish
Sep 6 at 8:16
Is $A=H$? If so, you may want to consistently use $H$ throughout.
â Meni Rosenfeld
Sep 6 at 12:52
@MeniRosenfeld Yes, of course. Gonna correct it. Thanks!
â Jose Brox
Sep 6 at 12:58
add a comment |Â
up vote
7
down vote
The characteristic polynomial of $H$ is $$X^3-4X^2+6X-3.$$
Therefore $$H^3=4H^2-6H+3I=H(4H-6I)+3I.$$
The element $(H^3)_23$ is then the product of the second row of $H$ times the third column of $4H-6I$ (plus nothing since $3I$ has element $(2,3)$ equal to $0$).
We need $18$ products and $12$ additions to get the polynomial, plus the expanding of terms. After that we need 1 addition (from the -6I), 3 multiplications (from the 4H), plus 3 multiplications and 2 additions from the second row-third column product.
But hey, this way is not dull!
1
Definitely not dull. I like creative usage of Cayley Hamilton! +1
â Babelfish
Sep 6 at 8:16
Is $A=H$? If so, you may want to consistently use $H$ throughout.
â Meni Rosenfeld
Sep 6 at 12:52
@MeniRosenfeld Yes, of course. Gonna correct it. Thanks!
â Jose Brox
Sep 6 at 12:58
add a comment |Â
up vote
7
down vote
up vote
7
down vote
The characteristic polynomial of $H$ is $$X^3-4X^2+6X-3.$$
Therefore $$H^3=4H^2-6H+3I=H(4H-6I)+3I.$$
The element $(H^3)_23$ is then the product of the second row of $H$ times the third column of $4H-6I$ (plus nothing since $3I$ has element $(2,3)$ equal to $0$).
We need $18$ products and $12$ additions to get the polynomial, plus the expanding of terms. After that we need 1 addition (from the -6I), 3 multiplications (from the 4H), plus 3 multiplications and 2 additions from the second row-third column product.
But hey, this way is not dull!
The characteristic polynomial of $H$ is $$X^3-4X^2+6X-3.$$
Therefore $$H^3=4H^2-6H+3I=H(4H-6I)+3I.$$
The element $(H^3)_23$ is then the product of the second row of $H$ times the third column of $4H-6I$ (plus nothing since $3I$ has element $(2,3)$ equal to $0$).
We need $18$ products and $12$ additions to get the polynomial, plus the expanding of terms. After that we need 1 addition (from the -6I), 3 multiplications (from the 4H), plus 3 multiplications and 2 additions from the second row-third column product.
But hey, this way is not dull!
edited Sep 6 at 12:59
answered Sep 6 at 8:11
Jose Brox
2,3631921
2,3631921
1
Definitely not dull. I like creative usage of Cayley Hamilton! +1
â Babelfish
Sep 6 at 8:16
Is $A=H$? If so, you may want to consistently use $H$ throughout.
â Meni Rosenfeld
Sep 6 at 12:52
@MeniRosenfeld Yes, of course. Gonna correct it. Thanks!
â Jose Brox
Sep 6 at 12:58
add a comment |Â
1
Definitely not dull. I like creative usage of Cayley Hamilton! +1
â Babelfish
Sep 6 at 8:16
Is $A=H$? If so, you may want to consistently use $H$ throughout.
â Meni Rosenfeld
Sep 6 at 12:52
@MeniRosenfeld Yes, of course. Gonna correct it. Thanks!
â Jose Brox
Sep 6 at 12:58
1
1
Definitely not dull. I like creative usage of Cayley Hamilton! +1
â Babelfish
Sep 6 at 8:16
Definitely not dull. I like creative usage of Cayley Hamilton! +1
â Babelfish
Sep 6 at 8:16
Is $A=H$? If so, you may want to consistently use $H$ throughout.
â Meni Rosenfeld
Sep 6 at 12:52
Is $A=H$? If so, you may want to consistently use $H$ throughout.
â Meni Rosenfeld
Sep 6 at 12:52
@MeniRosenfeld Yes, of course. Gonna correct it. Thanks!
â Jose Brox
Sep 6 at 12:58
@MeniRosenfeld Yes, of course. Gonna correct it. Thanks!
â Jose Brox
Sep 6 at 12:58
add a comment |Â
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2907167%2ffinding-a-certain-entry-in-a-matrix%23new-answer', 'question_page');
);
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password