turn off parts of features in a neural network? 2019 Community Moderator Election Results 2019 Moderator Election Q&A - QuestionnaireDebugging Neural Network for (Natural Language) TaggingTroubleshooting Neural Network ImplementationNeural Network for Multiple Output RegressionConvnet training error does not decreaseScalar input to neural network whose existence is conditionalchoosing sample points when approximating a function with a neural networkNeural network only converges when data cloud is close to 0How much neural network theory required to design one?Voice recognition with fourier transformation with audio input in pythonApplying ML to estimate parameters of an existing physical model
Are there existing rules/lore for MTG planeswalkers?
What is ls Largest Number Formed by only moving two sticks in 508?
What's the difference between using dependency injection with a container and using a service locator?
All ASCII characters with a given bit count
Determinant of a matrix with 2 equal rows
Is there an efficient way for synchronising audio events real-time with LEDs using an MCU?
What does こした mean?
How do I deal with an erroneously large refund?
What is /etc/mtab in Linux?
How did Elite on the NES work?
Putting Ant-Man on house arrest
Bright yellow or light yellow?
Suing a Police Officer Instead of the Police Department
What were wait-states, and why was it only an issue for PCs?
How would you suggest I follow up with coworkers about our deadline that's today?
Will I be more secure with my own router behind my ISP's router?
Is it appropriate to mention a relatable company blog post when you're asked about the company?
Why did Israel vote against lifting the American embargo on Cuba?
Where to find documentation for `whois` command options?
Variable does not exist: sObjectType (Task.sObjectType)
Could a cockatrice have parasitic embryos?
What's called a person who works as someone who puts products on shelves in stores?
Eigenvalues of the Laplacian of the directed De Bruijn graph
What is the ongoing value of the Kanban board to the developers as opposed to management
turn off parts of features in a neural network?
2019 Community Moderator Election Results
2019 Moderator Election Q&A - QuestionnaireDebugging Neural Network for (Natural Language) TaggingTroubleshooting Neural Network ImplementationNeural Network for Multiple Output RegressionConvnet training error does not decreaseScalar input to neural network whose existence is conditionalchoosing sample points when approximating a function with a neural networkNeural network only converges when data cloud is close to 0How much neural network theory required to design one?Voice recognition with fourier transformation with audio input in pythonApplying ML to estimate parameters of an existing physical model
$begingroup$
Suppose I have a neural network which accepts two sets of features as inputs and generates corresponding outputs, for instance, generate average final grade from: 1. working hours for N students in a class, 2. mid-term grade for N students in the same class. During training, the neural network is trained with many different classes. What I would like to during inference is to turn off one set of the features (i.e. feed only working hours as inputs) and predict outputs using the same neural network. Obviously, setting the mid-term grades to be all 0s would not be a good option. I wonder if anyone know what would be a good way to do this?
Thanks!
neural-network deep-learning
edited Sep 23 '18 at 17:44
Brian Spiering
4,2981129
asked Mar 26 '18 at 14:48
username123username123
1113
$endgroup$
bumped to the homepage by Community♦ 1 hour ago
This question has answers that may be good or bad; the system has marked it active so that they can be reviewed.
|
$begingroup$
Suppose I have a neural network which accepts two sets of features as inputs and generates corresponding outputs, for instance, generate average final grade from: 1. working hours for N students in a class, 2. mid-term grade for N students in the same class. During training, the neural network is trained with many different classes. What I would like to during inference is to turn off one set of the features (i.e. feed only working hours as inputs) and predict outputs using the same neural network. Obviously, setting the mid-term grades to be all 0s would not be a good option. I wonder if anyone know what would be a good way to do this?
Thanks!
neural-network deep-learning
edited Sep 23 '18 at 17:44
Brian Spiering
4,2981129
asked Mar 26 '18 at 14:48
username123username123
1113
$endgroup$
bumped to the homepage by Community♦ 1 hour ago
This question has answers that may be good or bad; the system has marked it active so that they can be reviewed.
$begingroup$
Why not only train of the features available at test time?
$endgroup$
– kbrose
May 25 '18 at 16:52
|
$begingroup$
Suppose I have a neural network which accepts two sets of features as inputs and generates corresponding outputs, for instance, generate average final grade from: 1. working hours for N students in a class, 2. mid-term grade for N students in the same class. During training, the neural network is trained with many different classes. What I would like to during inference is to turn off one set of the features (i.e. feed only working hours as inputs) and predict outputs using the same neural network. Obviously, setting the mid-term grades to be all 0s would not be a good option. I wonder if anyone know what would be a good way to do this?
Thanks!
neural-network deep-learning
edited Sep 23 '18 at 17:44
Brian Spiering
4,2981129
asked Mar 26 '18 at 14:48
username123username123
1113
$endgroup$
Suppose I have a neural network which accepts two sets of features as inputs and generates corresponding outputs, for instance, generate average final grade from: 1. working hours for N students in a class, 2. mid-term grade for N students in the same class. During training, the neural network is trained with many different classes. What I would like to during inference is to turn off one set of the features (i.e. feed only working hours as inputs) and predict outputs using the same neural network. Obviously, setting the mid-term grades to be all 0s would not be a good option. I wonder if anyone know what would be a good way to do this?
Thanks!
neural-network deep-learning
neural-network deep-learning
edited Sep 23 '18 at 17:44
Brian Spiering
4,2981129
asked Mar 26 '18 at 14:48
username123username123
1113
edited Sep 23 '18 at 17:44
Brian Spiering
4,2981129
asked Mar 26 '18 at 14:48
username123username123
1113
edited Sep 23 '18 at 17:44
Brian Spiering
4,2981129
edited Sep 23 '18 at 17:44
Brian Spiering
4,2981129
edited Sep 23 '18 at 17:44
Brian Spiering
4,2981129
4,2981129
asked Mar 26 '18 at 14:48
username123username123
1113
asked Mar 26 '18 at 14:48
username123username123
1113
asked Mar 26 '18 at 14:48
username123username123
1113
1113
bumped to the homepage by Community♦ 1 hour ago
This question has answers that may be good or bad; the system has marked it active so that they can be reviewed.
bumped to the homepage by Community♦ 1 hour ago
This question has answers that may be good or bad; the system has marked it active so that they can be reviewed.
$begingroup$
Why not only train of the features available at test time?
$endgroup$
– kbrose
May 25 '18 at 16:52
|
$begingroup$
Why not only train of the features available at test time?
$endgroup$
– kbrose
May 25 '18 at 16:52
$begingroup$
Why not only train of the features available at test time?
$endgroup$
– kbrose
May 25 '18 at 16:52
$begingroup$
Why not only train of the features available at test time?
$endgroup$
– kbrose
May 25 '18 at 16:52
|
1 Answer
1
active
oldest
votes
$begingroup$
As far as I know, you cannot do that.
First :
Obviously setting the mid-term grades to be all 0s would be be a good
option
No, actually that is a really bad option, neural networks do not understand magical parameters, meaning, if you put a 0 there it will think that the mid-term grades is 0, therefore it will probably give you a very low final grade prediction for you.
My advise, if you are very interested in doing that, create boolean factors, training inputs:
- working_hours
- mid-term_grade
- mid-term_grade_present
mid-term_grade_present should be a 1 when you have the mid-term_grade training data, and 0 when you do not have it. BTW, yes, generate training data without the mid-term_grade.
If you are working with non linear machine learning algorithms that should be enough, if you are using linear algorithms you should one hot encode the mid-term_grade_present and multiply it by the mid-term_grade, ending with something like this:
- working_hours
- mid-term_grade_present_1 * mid-term_grade
- mid-term_grade_present_0 * mid-term_grade
answered Mar 26 '18 at 15:10
KaileghKailegh
813
$endgroup$
$begingroup$
Sorry that was a typo, it should beObviously setting the mid-term grades to be all 0s would not be a good option.... I just updated it in the text. If I set mid-term_grade_present to be 0, what should I use as inputs, specifically what should be the data for mid-term_grade part? If it is the same withmid-term_grade_present=1case, then this boolean flag basically has no effect, if it is something else, what should it be?
$endgroup$
– username123
Mar 26 '18 at 15:23
$begingroup$
if you have enough data you could try taking 20% of it and setting mid-term_grade_present=0
$endgroup$
– Kailegh
Mar 26 '18 at 15:36
$begingroup$
If I setmid-term_grade_present=0I do not thinkmid_term_gradedata would be ignored by the neural network since: 1. it is already included as part of inputs, 2. it is correlated with outputs.
$endgroup$
– username123
Mar 26 '18 at 15:41
$begingroup$
1.-it is a neural network, it can learn non-linear relationships, and learn features interactions, trust me, it will learn that when mid-term_grade_present=0 it cant rely on mid-term_grade 2.- it is not correlated, you randomly pick a 20% of the data and remove the mid-term_grade info, no correlation in there, give a try, but I sincerely think it should work
$endgroup$
– Kailegh
Mar 26 '18 at 18:57
$begingroup$
did this idea finally work for you? or do we have to come up with something else?
$endgroup$
– Kailegh
Mar 27 '18 at 14:54
|
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
As far as I know, you cannot do that.
First :
Obviously setting the mid-term grades to be all 0s would be be a good
option
No, actually that is a really bad option, neural networks do not understand magical parameters, meaning, if you put a 0 there it will think that the mid-term grades is 0, therefore it will probably give you a very low final grade prediction for you.
My advise, if you are very interested in doing that, create boolean factors, training inputs:
- working_hours
- mid-term_grade
- mid-term_grade_present
mid-term_grade_present should be a 1 when you have the mid-term_grade training data, and 0 when you do not have it. BTW, yes, generate training data without the mid-term_grade.
If you are working with non linear machine learning algorithms that should be enough, if you are using linear algorithms you should one hot encode the mid-term_grade_present and multiply it by the mid-term_grade, ending with something like this:
- working_hours
- mid-term_grade_present_1 * mid-term_grade
- mid-term_grade_present_0 * mid-term_grade
answered Mar 26 '18 at 15:10
KaileghKailegh
813
$endgroup$
$begingroup$
Sorry that was a typo, it should beObviously setting the mid-term grades to be all 0s would not be a good option.... I just updated it in the text. If I set mid-term_grade_present to be 0, what should I use as inputs, specifically what should be the data for mid-term_grade part? If it is the same withmid-term_grade_present=1case, then this boolean flag basically has no effect, if it is something else, what should it be?
$endgroup$
– username123
Mar 26 '18 at 15:23
$begingroup$
if you have enough data you could try taking 20% of it and setting mid-term_grade_present=0
$endgroup$
– Kailegh
Mar 26 '18 at 15:36
$begingroup$
If I setmid-term_grade_present=0I do not thinkmid_term_gradedata would be ignored by the neural network since: 1. it is already included as part of inputs, 2. it is correlated with outputs.
$endgroup$
– username123
Mar 26 '18 at 15:41
$begingroup$
1.-it is a neural network, it can learn non-linear relationships, and learn features interactions, trust me, it will learn that when mid-term_grade_present=0 it cant rely on mid-term_grade 2.- it is not correlated, you randomly pick a 20% of the data and remove the mid-term_grade info, no correlation in there, give a try, but I sincerely think it should work
$endgroup$
– Kailegh
Mar 26 '18 at 18:57
$begingroup$
did this idea finally work for you? or do we have to come up with something else?
$endgroup$
– Kailegh
Mar 27 '18 at 14:54
|
$begingroup$
As far as I know, you cannot do that.
First :
Obviously setting the mid-term grades to be all 0s would be be a good
option
No, actually that is a really bad option, neural networks do not understand magical parameters, meaning, if you put a 0 there it will think that the mid-term grades is 0, therefore it will probably give you a very low final grade prediction for you.
My advise, if you are very interested in doing that, create boolean factors, training inputs:
- working_hours
- mid-term_grade
- mid-term_grade_present
mid-term_grade_present should be a 1 when you have the mid-term_grade training data, and 0 when you do not have it. BTW, yes, generate training data without the mid-term_grade.
If you are working with non linear machine learning algorithms that should be enough, if you are using linear algorithms you should one hot encode the mid-term_grade_present and multiply it by the mid-term_grade, ending with something like this:
- working_hours
- mid-term_grade_present_1 * mid-term_grade
- mid-term_grade_present_0 * mid-term_grade
answered Mar 26 '18 at 15:10
KaileghKailegh
813
$endgroup$
$begingroup$
Sorry that was a typo, it should beObviously setting the mid-term grades to be all 0s would not be a good option.... I just updated it in the text. If I set mid-term_grade_present to be 0, what should I use as inputs, specifically what should be the data for mid-term_grade part? If it is the same withmid-term_grade_present=1case, then this boolean flag basically has no effect, if it is something else, what should it be?
$endgroup$
– username123
Mar 26 '18 at 15:23
$begingroup$
if you have enough data you could try taking 20% of it and setting mid-term_grade_present=0
$endgroup$
– Kailegh
Mar 26 '18 at 15:36
$begingroup$
If I setmid-term_grade_present=0I do not thinkmid_term_gradedata would be ignored by the neural network since: 1. it is already included as part of inputs, 2. it is correlated with outputs.
$endgroup$
– username123
Mar 26 '18 at 15:41
$begingroup$
1.-it is a neural network, it can learn non-linear relationships, and learn features interactions, trust me, it will learn that when mid-term_grade_present=0 it cant rely on mid-term_grade 2.- it is not correlated, you randomly pick a 20% of the data and remove the mid-term_grade info, no correlation in there, give a try, but I sincerely think it should work
$endgroup$
– Kailegh
Mar 26 '18 at 18:57
$begingroup$
did this idea finally work for you? or do we have to come up with something else?
$endgroup$
– Kailegh
Mar 27 '18 at 14:54
|
$begingroup$
As far as I know, you cannot do that.
First :
Obviously setting the mid-term grades to be all 0s would be be a good
option
No, actually that is a really bad option, neural networks do not understand magical parameters, meaning, if you put a 0 there it will think that the mid-term grades is 0, therefore it will probably give you a very low final grade prediction for you.
My advise, if you are very interested in doing that, create boolean factors, training inputs:
- working_hours
- mid-term_grade
- mid-term_grade_present
mid-term_grade_present should be a 1 when you have the mid-term_grade training data, and 0 when you do not have it. BTW, yes, generate training data without the mid-term_grade.
If you are working with non linear machine learning algorithms that should be enough, if you are using linear algorithms you should one hot encode the mid-term_grade_present and multiply it by the mid-term_grade, ending with something like this:
- working_hours
- mid-term_grade_present_1 * mid-term_grade
- mid-term_grade_present_0 * mid-term_grade
answered Mar 26 '18 at 15:10
KaileghKailegh
813
$endgroup$
As far as I know, you cannot do that.
First :
Obviously setting the mid-term grades to be all 0s would be be a good
option
No, actually that is a really bad option, neural networks do not understand magical parameters, meaning, if you put a 0 there it will think that the mid-term grades is 0, therefore it will probably give you a very low final grade prediction for you.
My advise, if you are very interested in doing that, create boolean factors, training inputs:
- working_hours
- mid-term_grade
- mid-term_grade_present
mid-term_grade_present should be a 1 when you have the mid-term_grade training data, and 0 when you do not have it. BTW, yes, generate training data without the mid-term_grade.
If you are working with non linear machine learning algorithms that should be enough, if you are using linear algorithms you should one hot encode the mid-term_grade_present and multiply it by the mid-term_grade, ending with something like this:
- working_hours
- mid-term_grade_present_1 * mid-term_grade
- mid-term_grade_present_0 * mid-term_grade
answered Mar 26 '18 at 15:10
KaileghKailegh
813
answered Mar 26 '18 at 15:10
KaileghKailegh
813
answered Mar 26 '18 at 15:10
KaileghKailegh
813
answered Mar 26 '18 at 15:10
KaileghKailegh
813
813
$begingroup$
Sorry that was a typo, it should beObviously setting the mid-term grades to be all 0s would not be a good option.... I just updated it in the text. If I set mid-term_grade_present to be 0, what should I use as inputs, specifically what should be the data for mid-term_grade part? If it is the same withmid-term_grade_present=1case, then this boolean flag basically has no effect, if it is something else, what should it be?
$endgroup$
– username123
Mar 26 '18 at 15:23
$begingroup$
if you have enough data you could try taking 20% of it and setting mid-term_grade_present=0
$endgroup$
– Kailegh
Mar 26 '18 at 15:36
$begingroup$
If I setmid-term_grade_present=0I do not thinkmid_term_gradedata would be ignored by the neural network since: 1. it is already included as part of inputs, 2. it is correlated with outputs.
$endgroup$
– username123
Mar 26 '18 at 15:41
$begingroup$
1.-it is a neural network, it can learn non-linear relationships, and learn features interactions, trust me, it will learn that when mid-term_grade_present=0 it cant rely on mid-term_grade 2.- it is not correlated, you randomly pick a 20% of the data and remove the mid-term_grade info, no correlation in there, give a try, but I sincerely think it should work
$endgroup$
– Kailegh
Mar 26 '18 at 18:57
$begingroup$
did this idea finally work for you? or do we have to come up with something else?
$endgroup$
– Kailegh
Mar 27 '18 at 14:54
|
$begingroup$
Sorry that was a typo, it should beObviously setting the mid-term grades to be all 0s would not be a good option.... I just updated it in the text. If I set mid-term_grade_present to be 0, what should I use as inputs, specifically what should be the data for mid-term_grade part? If it is the same withmid-term_grade_present=1case, then this boolean flag basically has no effect, if it is something else, what should it be?
$endgroup$
– username123
Mar 26 '18 at 15:23
$begingroup$
if you have enough data you could try taking 20% of it and setting mid-term_grade_present=0
$endgroup$
– Kailegh
Mar 26 '18 at 15:36
$begingroup$
If I setmid-term_grade_present=0I do not thinkmid_term_gradedata would be ignored by the neural network since: 1. it is already included as part of inputs, 2. it is correlated with outputs.
$endgroup$
– username123
Mar 26 '18 at 15:41
$begingroup$
1.-it is a neural network, it can learn non-linear relationships, and learn features interactions, trust me, it will learn that when mid-term_grade_present=0 it cant rely on mid-term_grade 2.- it is not correlated, you randomly pick a 20% of the data and remove the mid-term_grade info, no correlation in there, give a try, but I sincerely think it should work
$endgroup$
– Kailegh
Mar 26 '18 at 18:57
$begingroup$
did this idea finally work for you? or do we have to come up with something else?
$endgroup$
– Kailegh
Mar 27 '18 at 14:54
$begingroup$
Sorry that was a typo, it should be
Obviously setting the mid-term grades to be all 0s would not be a good option.... I just updated it in the text. If I set mid-term_grade_present to be 0, what should I use as inputs, specifically what should be the data for mid-term_grade part? If it is the same with mid-term_grade_present=1 case, then this boolean flag basically has no effect, if it is something else, what should it be?$endgroup$
– username123
Mar 26 '18 at 15:23
$begingroup$
Sorry that was a typo, it should be
Obviously setting the mid-term grades to be all 0s would not be a good option.... I just updated it in the text. If I set mid-term_grade_present to be 0, what should I use as inputs, specifically what should be the data for mid-term_grade part? If it is the same with mid-term_grade_present=1 case, then this boolean flag basically has no effect, if it is something else, what should it be?$endgroup$
– username123
Mar 26 '18 at 15:23
$begingroup$
if you have enough data you could try taking 20% of it and setting mid-term_grade_present=0
$endgroup$
– Kailegh
Mar 26 '18 at 15:36
$begingroup$
if you have enough data you could try taking 20% of it and setting mid-term_grade_present=0
$endgroup$
– Kailegh
Mar 26 '18 at 15:36
$begingroup$
If I set
mid-term_grade_present=0 I do not think mid_term_grade data would be ignored by the neural network since: 1. it is already included as part of inputs, 2. it is correlated with outputs.$endgroup$
– username123
Mar 26 '18 at 15:41
$begingroup$
If I set
mid-term_grade_present=0 I do not think mid_term_grade data would be ignored by the neural network since: 1. it is already included as part of inputs, 2. it is correlated with outputs.$endgroup$
– username123
Mar 26 '18 at 15:41
$begingroup$
1.-it is a neural network, it can learn non-linear relationships, and learn features interactions, trust me, it will learn that when mid-term_grade_present=0 it cant rely on mid-term_grade 2.- it is not correlated, you randomly pick a 20% of the data and remove the mid-term_grade info, no correlation in there, give a try, but I sincerely think it should work
$endgroup$
– Kailegh
Mar 26 '18 at 18:57
$begingroup$
1.-it is a neural network, it can learn non-linear relationships, and learn features interactions, trust me, it will learn that when mid-term_grade_present=0 it cant rely on mid-term_grade 2.- it is not correlated, you randomly pick a 20% of the data and remove the mid-term_grade info, no correlation in there, give a try, but I sincerely think it should work
$endgroup$
– Kailegh
Mar 26 '18 at 18:57
$begingroup$
did this idea finally work for you? or do we have to come up with something else?
$endgroup$
– Kailegh
Mar 27 '18 at 14:54
$begingroup$
did this idea finally work for you? or do we have to come up with something else?
$endgroup$
– Kailegh
Mar 27 '18 at 14:54
|
$begingroup$
Why not only train of the features available at test time?
$endgroup$
– kbrose
May 25 '18 at 16:52