How can NN be thought of as p(x|z) in VAEs? Announcing the arrival of Valued Associate #679: Cesar Manara Planned maintenance scheduled April 23, 2019 at 23:30 UTC (7:30 pm US/Eastern) 2019 Moderator Election Q&A - Questionnaire 2019 Community Moderator Election ResultsOnline Variational AutoencoderTips and tricks for designing time-series variational autoencodersHow do I provide input and output for such a network structure in kerasUnsupervised Anomaly Detection in Imageswhy is mse training drastically different from the begining of each training with Encoder-DecoderTraining of Region Proposal Network (RPN)Keras: Softmax output into embedding layerKeras input dimensions for a MLPEncoder Decoder networks with varying image sizesHow to build this simple generative model?

My admission is revoked after accepting the admission offer

Is there a possibility to generate a list dynamically in Latex?

What were wait-states, and why was it only an issue for PCs?

What is the evidence that custom checks in Northern Ireland are going to result in violence?

Protagonist's race is hidden - should I reveal it?

How to translate "red flag" into Spanish?

Co-worker works way more than he should

How was Lagrange appointed professor of mathematics so early?

Was there ever a LEGO store in Miami International Airport?

What's parked in Mil Moscow helicopter plant?

Like totally amazing interchangeable sister outfit accessory swapping or whatever

Variable does not exist: sObjectType (Task.sObjectType)

Why aren't road bicycle wheels tiny?

What is the ongoing value of the Kanban board to the developers as opposed to management

Raising a bilingual kid. When should we introduce the majority language?

What does こした mean?

Why doesn't the university give past final exams' answers?

Does Prince Arnaud cause someone holding the Princess to lose?

Israeli soda type drink

RIP Packet Format

What is the purpose of the side handle on a hand ("eggbeater") drill?

Is there a way to fake a method response using Mock or Stubs?

How to begin with a paragraph in latex

Could a cockatrice have parasitic embryos?



How can NN be thought of as p(x|z) in VAEs?



Announcing the arrival of Valued Associate #679: Cesar Manara
Planned maintenance scheduled April 23, 2019 at 23:30 UTC (7:30 pm US/Eastern)
2019 Moderator Election Q&A - Questionnaire
2019 Community Moderator Election ResultsOnline Variational AutoencoderTips and tricks for designing time-series variational autoencodersHow do I provide input and output for such a network structure in kerasUnsupervised Anomaly Detection in Imageswhy is mse training drastically different from the begining of each training with Encoder-DecoderTraining of Region Proposal Network (RPN)Keras: Softmax output into embedding layerKeras input dimensions for a MLPEncoder Decoder networks with varying image sizesHow to build this simple generative model?










0












$begingroup$


So, you can feed random input (z) to a NN and make the output random, but that will be just f(p(z)), where f(.) is the deterministic NN. The other way is to think of the NN output as parameteres of a certain distribution and as such, the NN is P(x|z). BUT, that's not happening, the decoder of a VAE is producing an image (x), and it is not random! so, for a given z, it will produce the same x, all the time. Why do they think of it as P(x|z) in the derivations of VAEs??










share|improve this question







New contributor




Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.







$endgroup$
















    0












    $begingroup$


    So, you can feed random input (z) to a NN and make the output random, but that will be just f(p(z)), where f(.) is the deterministic NN. The other way is to think of the NN output as parameteres of a certain distribution and as such, the NN is P(x|z). BUT, that's not happening, the decoder of a VAE is producing an image (x), and it is not random! so, for a given z, it will produce the same x, all the time. Why do they think of it as P(x|z) in the derivations of VAEs??










    share|improve this question







    New contributor




    Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
    Check out our Code of Conduct.







    $endgroup$














      0












      0








      0





      $begingroup$


      So, you can feed random input (z) to a NN and make the output random, but that will be just f(p(z)), where f(.) is the deterministic NN. The other way is to think of the NN output as parameteres of a certain distribution and as such, the NN is P(x|z). BUT, that's not happening, the decoder of a VAE is producing an image (x), and it is not random! so, for a given z, it will produce the same x, all the time. Why do they think of it as P(x|z) in the derivations of VAEs??










      share|improve this question







      New contributor




      Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
      Check out our Code of Conduct.







      $endgroup$




      So, you can feed random input (z) to a NN and make the output random, but that will be just f(p(z)), where f(.) is the deterministic NN. The other way is to think of the NN output as parameteres of a certain distribution and as such, the NN is P(x|z). BUT, that's not happening, the decoder of a VAE is producing an image (x), and it is not random! so, for a given z, it will produce the same x, all the time. Why do they think of it as P(x|z) in the derivations of VAEs??







      machine-learning deep-learning keras






      share|improve this question







      New contributor




      Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
      Check out our Code of Conduct.











      share|improve this question







      New contributor




      Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
      Check out our Code of Conduct.









      share|improve this question




      share|improve this question






      New contributor




      Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
      Check out our Code of Conduct.









      asked 16 mins ago









      Alex DeftAlex Deft

      1




      1




      New contributor




      Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
      Check out our Code of Conduct.





      New contributor





      Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
      Check out our Code of Conduct.






      Alex Deft is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
      Check out our Code of Conduct.




















          0






          active

          oldest

          votes












          Your Answer








          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
          );



          );






          Alex Deft is a new contributor. Be nice, and check out our Code of Conduct.









          draft saved

          draft discarded


















          StackExchange.ready(
          function ()
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fdatascience.stackexchange.com%2fquestions%2f50810%2fhow-can-nn-be-thought-of-as-pxz-in-vaes%23new-answer', 'question_page');

          );

          Post as a guest















          Required, but never shown

























          0






          active

          oldest

          votes








          0






          active

          oldest

          votes









          active

          oldest

          votes






          active

          oldest

          votes








          Alex Deft is a new contributor. Be nice, and check out our Code of Conduct.









          draft saved

          draft discarded


















          Alex Deft is a new contributor. Be nice, and check out our Code of Conduct.












          Alex Deft is a new contributor. Be nice, and check out our Code of Conduct.











          Alex Deft is a new contributor. Be nice, and check out our Code of Conduct.














          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.




          draft saved


          draft discarded














          StackExchange.ready(
          function ()
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fdatascience.stackexchange.com%2fquestions%2f50810%2fhow-can-nn-be-thought-of-as-pxz-in-vaes%23new-answer', 'question_page');

          );

          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







          Popular posts from this blog

          ValueError: Error when checking input: expected conv2d_13_input to have shape (3, 150, 150) but got array with shape (150, 150, 3)2019 Community Moderator ElectionError when checking : expected dense_1_input to have shape (None, 5) but got array with shape (200, 1)Error 'Expected 2D array, got 1D array instead:'ValueError: Error when checking input: expected lstm_41_input to have 3 dimensions, but got array with shape (40000,100)ValueError: Error when checking target: expected dense_1 to have shape (7,) but got array with shape (1,)ValueError: Error when checking target: expected dense_2 to have shape (1,) but got array with shape (0,)Keras exception: ValueError: Error when checking input: expected conv2d_1_input to have shape (150, 150, 3) but got array with shape (256, 256, 3)Steps taking too long to completewhen checking input: expected dense_1_input to have shape (13328,) but got array with shape (317,)ValueError: Error when checking target: expected dense_3 to have shape (None, 1) but got array with shape (7715, 40000)Keras exception: Error when checking input: expected dense_input to have shape (2,) but got array with shape (1,)

          Ружовы пелікан Змест Знешні выгляд | Пашырэнне | Асаблівасці біялогіі | Літаратура | НавігацыяДагледжаная версіяправерана1 зменаДагледжаная версіяправерана1 змена/ 22697590 Сістэматыкана ВіківідахВыявына Вікісховішчы174693363011049382

          Illegal assignment from SObject to ContactFetching String, Id from Map - Illegal Assignment Id to Field / ObjectError: Compile Error: Illegal assignment from String to BooleanError: List has no rows for assignment to SObjectError on Test Class - System.QueryException: List has no rows for assignment to SObjectRemote action problemDML requires SObject or SObject list type error“Illegal assignment from List to List”Test Class Fail: Batch Class: System.QueryException: List has no rows for assignment to SObjectMapping to a user'List has no rows for assignment to SObject' Mystery