What activation function should I use for a specific regression problem?2019 Community Moderator ElectionHow flexible is the link between objective function and output layer activation function?Technology stack for linear regression on (not so) large datasetParameterization regression of rotation angleAlternatives to linear activation function in regression tasks to limit the outputWhat is the Time Complexity of Linear Regression?Activation function vs Squashing functionUsing Keras to Predict a Function Following a Normal Distributionbest activation function for ensemble?Regression with -1,1 target range - Should we use a tanh activation in the last 1 unit dense layer?Images Score Regression only regresses to the average of the target values
How to verify if g is a generator for p?
What are the ramifications of creating a homebrew world without an Astral Plane?
quarter to five p.m
Can I convert a rim brake wheel to a disc brake wheel?
Do I need a multiple entry visa for a trip UK -> Sweden -> UK?
Hide Select Output from T-SQL
How can I use the arrow sign in my bash prompt?
Opposite of a diet
Valid Badminton Score?
Was the picture area of a CRT a parallelogram (instead of a true rectangle)?
How will losing mobility of one hand affect my career as a programmer?
How does a character multiclassing into warlock get a focus?
Is there a problem with hiding "forgot password" until it's needed?
How do I define a right arrow with bar in LaTeX?
Should my PhD thesis be submitted under my legal name?
Cynical novel that describes an America ruled by the media, arms manufacturers, and ethnic figureheads
Efficiently merge handle parallel feature branches in SFDX
Trouble understanding overseas colleagues
Why is delta-v is the most useful quantity for planning space travel?
What's a natural way to say that someone works somewhere (for a job)?
Is a roofing delivery truck likely to crack my driveway slab?
How do I rename a LINUX host without needing to reboot for the rename to take effect?
Have I saved too much for retirement so far?
How to prove that the query oracle is unitary?
What activation function should I use for a specific regression problem?
2019 Community Moderator ElectionHow flexible is the link between objective function and output layer activation function?Technology stack for linear regression on (not so) large datasetParameterization regression of rotation angleAlternatives to linear activation function in regression tasks to limit the outputWhat is the Time Complexity of Linear Regression?Activation function vs Squashing functionUsing Keras to Predict a Function Following a Normal Distributionbest activation function for ensemble?Regression with -1,1 target range - Should we use a tanh activation in the last 1 unit dense layer?Images Score Regression only regresses to the average of the target values
$begingroup$
Which is better for regression problems create a neural net with tanh/sigmoid and exp(like) activations or ReLU and linear? Standard is to use ReLU but it's brute force solution that requires certain net size and I would like to avoid creating a very big net, also sigmoid is much more prefered but in my case regression will output values from range (0, 1e7)... maybe also sigmoid net with linear head will work? I am curious about your take on the subject.
machine-learning neural-network deep-learning regression activation-function
edited Mar 21 at 21:51
Media
7,42262162
asked Mar 21 at 18:14
questerquester
384
$endgroup$
add a comment |
$begingroup$
Which is better for regression problems create a neural net with tanh/sigmoid and exp(like) activations or ReLU and linear? Standard is to use ReLU but it's brute force solution that requires certain net size and I would like to avoid creating a very big net, also sigmoid is much more prefered but in my case regression will output values from range (0, 1e7)... maybe also sigmoid net with linear head will work? I am curious about your take on the subject.
machine-learning neural-network deep-learning regression activation-function
edited Mar 21 at 21:51
Media
7,42262162
asked Mar 21 at 18:14
questerquester
384
$endgroup$
add a comment |
$begingroup$
Which is better for regression problems create a neural net with tanh/sigmoid and exp(like) activations or ReLU and linear? Standard is to use ReLU but it's brute force solution that requires certain net size and I would like to avoid creating a very big net, also sigmoid is much more prefered but in my case regression will output values from range (0, 1e7)... maybe also sigmoid net with linear head will work? I am curious about your take on the subject.
machine-learning neural-network deep-learning regression activation-function
edited Mar 21 at 21:51
Media
7,42262162
asked Mar 21 at 18:14
questerquester
384
$endgroup$
Which is better for regression problems create a neural net with tanh/sigmoid and exp(like) activations or ReLU and linear? Standard is to use ReLU but it's brute force solution that requires certain net size and I would like to avoid creating a very big net, also sigmoid is much more prefered but in my case regression will output values from range (0, 1e7)... maybe also sigmoid net with linear head will work? I am curious about your take on the subject.
machine-learning neural-network deep-learning regression activation-function
machine-learning neural-network deep-learning regression activation-function
edited Mar 21 at 21:51
Media
7,42262162
asked Mar 21 at 18:14
questerquester
384
edited Mar 21 at 21:51
Media
7,42262162
asked Mar 21 at 18:14
questerquester
384
edited Mar 21 at 21:51
Media
7,42262162
edited Mar 21 at 21:51
Media
7,42262162
edited Mar 21 at 21:51
Media
7,42262162
7,42262162
asked Mar 21 at 18:14
questerquester
384
asked Mar 21 at 18:14
questerquester
384
asked Mar 21 at 18:14
questerquester
384
384
add a comment |
add a comment |
2 Answers
2
active
oldest
votes
$begingroup$
There are two points that have to be considered.
- Take care of the output of your network. If that's a Real number and can take any value, you have to use linear activation as the output.
- The inner activations highly depend on your task and the size of the network that you use. What I'm going to tell you is based on experience. If you don't have a very deep network, $tanh$ and $ReLU$ may not differ very much in convergence time. If you're using very deep networks, don't use $tahn$ at all. $ReLU$ is also not recommended in some contexts. You can employ $PReLU$ in very deep networks. It does not add too many parameters to learn. You can also use $leaky-ReLU$ in order to avoid dying ReLU problem which may occur.
Finally, about the other nonlinearity that you've referred; try not to use $Sigmoid$ due to that fact that it's expected value is not equal to zero but half. It's a bit statistical stuff, but you can consider it's roughly hard for a network to learn shifted weights.
answered Mar 21 at 21:49
MediaMedia
7,42262162
$endgroup$
add a comment |
$begingroup$
The issue with sigmoid and tanh activations is that their gradients saturate for extreme values of their arguments. This may occur if you do not normalize your inputs. If the learned weights of the unit are such that the gradient of its activation is close to zero, it will take longer for any updates to be reflected in the unit's weights. A final layer with no non-linearity will help you scale up your hidden layers' outputs.
In the end, the performance is application specific. You should try out both kinds of activations on a subset of your data and see which performs better.
answered Mar 21 at 18:25
hazrmardhazrmard
1364
$endgroup$
add a comment |
Your Answer
StackExchange.ifUsing("editor", function ()
return StackExchange.using("mathjaxEditing", function ()
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix)
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
);
);
, "mathjax-editing");
StackExchange.ready(function()
var channelOptions =
tags: "".split(" "),
id: "557"
;
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function()
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled)
StackExchange.using("snippets", function()
createEditor();
);
else
createEditor();
);
function createEditor()
StackExchange.prepareEditor(
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: false,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: null,
bindNavPrevention: true,
postfix: "",
imageUploader:
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
,
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
);
);
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
Required, but never shown
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fdatascience.stackexchange.com%2fquestions%2f47751%2fwhat-activation-function-should-i-use-for-a-specific-regression-problem%23new-answer', 'question_page');
);
Post as a guest
Required, but never shown
2 Answers
2
active
oldest
votes
2 Answers
2
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
There are two points that have to be considered.
- Take care of the output of your network. If that's a Real number and can take any value, you have to use linear activation as the output.
- The inner activations highly depend on your task and the size of the network that you use. What I'm going to tell you is based on experience. If you don't have a very deep network, $tanh$ and $ReLU$ may not differ very much in convergence time. If you're using very deep networks, don't use $tahn$ at all. $ReLU$ is also not recommended in some contexts. You can employ $PReLU$ in very deep networks. It does not add too many parameters to learn. You can also use $leaky-ReLU$ in order to avoid dying ReLU problem which may occur.
Finally, about the other nonlinearity that you've referred; try not to use $Sigmoid$ due to that fact that it's expected value is not equal to zero but half. It's a bit statistical stuff, but you can consider it's roughly hard for a network to learn shifted weights.
answered Mar 21 at 21:49
MediaMedia
7,42262162
$endgroup$
add a comment |
$begingroup$
There are two points that have to be considered.
- Take care of the output of your network. If that's a Real number and can take any value, you have to use linear activation as the output.
- The inner activations highly depend on your task and the size of the network that you use. What I'm going to tell you is based on experience. If you don't have a very deep network, $tanh$ and $ReLU$ may not differ very much in convergence time. If you're using very deep networks, don't use $tahn$ at all. $ReLU$ is also not recommended in some contexts. You can employ $PReLU$ in very deep networks. It does not add too many parameters to learn. You can also use $leaky-ReLU$ in order to avoid dying ReLU problem which may occur.
Finally, about the other nonlinearity that you've referred; try not to use $Sigmoid$ due to that fact that it's expected value is not equal to zero but half. It's a bit statistical stuff, but you can consider it's roughly hard for a network to learn shifted weights.
answered Mar 21 at 21:49
MediaMedia
7,42262162
$endgroup$
add a comment |
$begingroup$
There are two points that have to be considered.
- Take care of the output of your network. If that's a Real number and can take any value, you have to use linear activation as the output.
- The inner activations highly depend on your task and the size of the network that you use. What I'm going to tell you is based on experience. If you don't have a very deep network, $tanh$ and $ReLU$ may not differ very much in convergence time. If you're using very deep networks, don't use $tahn$ at all. $ReLU$ is also not recommended in some contexts. You can employ $PReLU$ in very deep networks. It does not add too many parameters to learn. You can also use $leaky-ReLU$ in order to avoid dying ReLU problem which may occur.
Finally, about the other nonlinearity that you've referred; try not to use $Sigmoid$ due to that fact that it's expected value is not equal to zero but half. It's a bit statistical stuff, but you can consider it's roughly hard for a network to learn shifted weights.
answered Mar 21 at 21:49
MediaMedia
7,42262162
$endgroup$
There are two points that have to be considered.
- Take care of the output of your network. If that's a Real number and can take any value, you have to use linear activation as the output.
- The inner activations highly depend on your task and the size of the network that you use. What I'm going to tell you is based on experience. If you don't have a very deep network, $tanh$ and $ReLU$ may not differ very much in convergence time. If you're using very deep networks, don't use $tahn$ at all. $ReLU$ is also not recommended in some contexts. You can employ $PReLU$ in very deep networks. It does not add too many parameters to learn. You can also use $leaky-ReLU$ in order to avoid dying ReLU problem which may occur.
Finally, about the other nonlinearity that you've referred; try not to use $Sigmoid$ due to that fact that it's expected value is not equal to zero but half. It's a bit statistical stuff, but you can consider it's roughly hard for a network to learn shifted weights.
answered Mar 21 at 21:49
MediaMedia
7,42262162
edited Mar 21 at 22:00
answered Mar 21 at 21:49
MediaMedia
7,42262162
answered Mar 21 at 21:49
MediaMedia
7,42262162
answered Mar 21 at 21:49
MediaMedia
7,42262162
7,42262162
add a comment |
add a comment |
$begingroup$
The issue with sigmoid and tanh activations is that their gradients saturate for extreme values of their arguments. This may occur if you do not normalize your inputs. If the learned weights of the unit are such that the gradient of its activation is close to zero, it will take longer for any updates to be reflected in the unit's weights. A final layer with no non-linearity will help you scale up your hidden layers' outputs.
In the end, the performance is application specific. You should try out both kinds of activations on a subset of your data and see which performs better.
answered Mar 21 at 18:25
hazrmardhazrmard
1364
$endgroup$
add a comment |
$begingroup$
The issue with sigmoid and tanh activations is that their gradients saturate for extreme values of their arguments. This may occur if you do not normalize your inputs. If the learned weights of the unit are such that the gradient of its activation is close to zero, it will take longer for any updates to be reflected in the unit's weights. A final layer with no non-linearity will help you scale up your hidden layers' outputs.
In the end, the performance is application specific. You should try out both kinds of activations on a subset of your data and see which performs better.
answered Mar 21 at 18:25
hazrmardhazrmard
1364
$endgroup$
add a comment |
$begingroup$
The issue with sigmoid and tanh activations is that their gradients saturate for extreme values of their arguments. This may occur if you do not normalize your inputs. If the learned weights of the unit are such that the gradient of its activation is close to zero, it will take longer for any updates to be reflected in the unit's weights. A final layer with no non-linearity will help you scale up your hidden layers' outputs.
In the end, the performance is application specific. You should try out both kinds of activations on a subset of your data and see which performs better.
answered Mar 21 at 18:25
hazrmardhazrmard
1364
$endgroup$
The issue with sigmoid and tanh activations is that their gradients saturate for extreme values of their arguments. This may occur if you do not normalize your inputs. If the learned weights of the unit are such that the gradient of its activation is close to zero, it will take longer for any updates to be reflected in the unit's weights. A final layer with no non-linearity will help you scale up your hidden layers' outputs.
In the end, the performance is application specific. You should try out both kinds of activations on a subset of your data and see which performs better.
answered Mar 21 at 18:25
hazrmardhazrmard
1364
answered Mar 21 at 18:25
hazrmardhazrmard
1364
answered Mar 21 at 18:25
hazrmardhazrmard
1364
answered Mar 21 at 18:25
hazrmardhazrmard
1364
1364
add a comment |
add a comment |
Thanks for contributing an answer to Data Science Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
Use MathJax to format equations. MathJax reference.
To learn more, see our tips on writing great answers.
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
Required, but never shown
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fdatascience.stackexchange.com%2fquestions%2f47751%2fwhat-activation-function-should-i-use-for-a-specific-regression-problem%23new-answer', 'question_page');
);
Post as a guest
Required, but never shown
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
Required, but never shown
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
Required, but never shown
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
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
