Trjtdtk

How do I deal with a coworker that keeps asking to make small superficial changes to a report, and it is seriously triggering my anxiety?

Is there any official lore on the Far Realm?

Why does nature favour the Laplacian?

What does the integral of a function times a function of a random variable represent, conceptually?

How would 10 generations of living underground change the human body?

How come there are so many candidates for the 2020 Democratic party presidential nomination?

Why did some of my point & shoot film photos come back with one third light white or orange?

Why was the Spitfire's elliptical wing almost uncopied by other aircraft of World War 2?

Elements that can bond to themselves?

Coordinate my way to the name of the (video) game

bldc motor, esc and battery draw, nominal vs peak

Why does Mind Blank stop the Feeblemind spell?

Can an Area of Effect spell cast outside a Prismatic Wall extend inside it?

What is the most expensive material in the world that could be used to create Pun-Pun's lute?

Was there a Viking Exchange as well as a Columbian one?

How exactly does Hawking radiation decrease the mass of black holes?

Contradiction proof for inequality of P and NP?

Why did C use the -> operator instead of reusing the . operator?

On The Origin of Dissonant Chords

Did the BCPL programming language support floats?

Dynamic SOQL query relationship with field visibility for Users

a sore throat vs a strep throat vs strep throat

Is Diceware more secure than a long passphrase?

Is there a way to generate a list of distinct numbers such that no two subsets ever have an equal sum?

How does on test regression for a subspace or matrix factorization?

How does the test data gets collected?matrix factorization?How exactly does matrix factorization help with collaborative filteringKernelized Probabilistic Matrix Factorization - Implementation?Does a matrix factorization recommendation engine use user/item related features?Regularization term in Matrix FactorizationMatrix Factorization for Recommender SystemsHow dot product limits expressiveness and leads to sub-optimal solutions in Matrix Factorization?Finding unobserved ratings using matrix factorizationDoes cardinality of ratings column affect performance of matrix factorization based collaborative filtering?

0

$begingroup$

I've recently been reading a lot of papers and watching a lot of videos on both subspace learning, and matrix factorization. One thing is particularly eluding me though - how does any of this get tested?

Let's take, straight from Wikipedia, Matrix Factorization Non-Negative.

$$V = WH$$

So, you have a data matrix $V$. Your goal is to learn components $W$ and $H$, which when multiplied together, give a good approximation of $V$. This can be done by minimizing over $W$, $H$

$$| V - WH |$$

That seems fine so far. My problem, theoretically, is understanding when we want to apply this to a problem, like say Regression.

If you wanted to minimize:

$$Y - WH*B$$

How do you do this with a test point? I get confused here, because if we had, say a 100-user test set with 10 features. Then we do a 90/10 split, we get a size of $W*H$ that is different than the size of our test data.

Do people just plug the test data in directly when testing, in place of $W*H$, and just rely on those learned weights $B$?

regression linear-regression matrix-factorisation matrix

share|improve this question

edited Apr 8 at 0:19

Stephen Rauch♦

1,52551330

asked Apr 7 at 23:23

JibrilJibril

1111

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I think your analogy is wrong. We don't do any matrix factorization in linear regression. We try to find $W$ in $Y approx WX$ where $Y$ and $X$ are given. In matrix factorization $Y approx WH$, we want to find $W$ and $H$ and only $Y$ is given.
  $endgroup$
  – Esmailian
  Apr 8 at 11:12
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Appreciate the reply. Is there a reason, other than the problem above, why we can't do matrix factorization for linear regression? I was trying to think of use-cases where you want to complete a matrix with missing data, but traditional columnwise or rowwise imputation may not make sense.
  $endgroup$
  – Jibril
  Apr 8 at 11:16
  

  
  
  
  

add a comment | 

0

$begingroup$

I've recently been reading a lot of papers and watching a lot of videos on both subspace learning, and matrix factorization. One thing is particularly eluding me though - how does any of this get tested?

Let's take, straight from Wikipedia, Matrix Factorization Non-Negative.

$$V = WH$$

So, you have a data matrix $V$. Your goal is to learn components $W$ and $H$, which when multiplied together, give a good approximation of $V$. This can be done by minimizing over $W$, $H$

$$| V - WH |$$

That seems fine so far. My problem, theoretically, is understanding when we want to apply this to a problem, like say Regression.

If you wanted to minimize:

$$Y - WH*B$$

How do you do this with a test point? I get confused here, because if we had, say a 100-user test set with 10 features. Then we do a 90/10 split, we get a size of $W*H$ that is different than the size of our test data.

Do people just plug the test data in directly when testing, in place of $W*H$, and just rely on those learned weights $B$?

regression linear-regression matrix-factorisation matrix

share|improve this question

edited Apr 8 at 0:19

Stephen Rauch♦

1,52551330

asked Apr 7 at 23:23

JibrilJibril

1111

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I think your analogy is wrong. We don't do any matrix factorization in linear regression. We try to find $W$ in $Y approx WX$ where $Y$ and $X$ are given. In matrix factorization $Y approx WH$, we want to find $W$ and $H$ and only $Y$ is given.
  $endgroup$
  – Esmailian
  Apr 8 at 11:12
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Appreciate the reply. Is there a reason, other than the problem above, why we can't do matrix factorization for linear regression? I was trying to think of use-cases where you want to complete a matrix with missing data, but traditional columnwise or rowwise imputation may not make sense.
  $endgroup$
  – Jibril
  Apr 8 at 11:16
  

  
  
  
  

add a comment | 

$begingroup$

I've recently been reading a lot of papers and watching a lot of videos on both subspace learning, and matrix factorization. One thing is particularly eluding me though - how does any of this get tested?

Let's take, straight from Wikipedia, Matrix Factorization Non-Negative.

$$V = WH$$

So, you have a data matrix $V$. Your goal is to learn components $W$ and $H$, which when multiplied together, give a good approximation of $V$. This can be done by minimizing over $W$, $H$

$$| V - WH |$$

That seems fine so far. My problem, theoretically, is understanding when we want to apply this to a problem, like say Regression.

If you wanted to minimize:

$$Y - WH*B$$

How do you do this with a test point? I get confused here, because if we had, say a 100-user test set with 10 features. Then we do a 90/10 split, we get a size of $W*H$ that is different than the size of our test data.

Do people just plug the test data in directly when testing, in place of $W*H$, and just rely on those learned weights $B$?

regression linear-regression matrix-factorisation matrix

share|improve this question

edited Apr 8 at 0:19

Stephen Rauch♦

1,52551330

asked Apr 7 at 23:23

JibrilJibril

1111

$endgroup$

share|improve this question

edited Apr 8 at 0:19

Stephen Rauch♦

1,52551330

asked Apr 7 at 23:23

JibrilJibril

1111

share|improve this question

edited Apr 8 at 0:19

Stephen Rauch♦

1,52551330

asked Apr 7 at 23:23

JibrilJibril

1111

edited Apr 8 at 0:19

Stephen Rauch♦

1,52551330

edited Apr 8 at 0:19

Stephen Rauch♦

1,52551330

asked Apr 7 at 23:23

JibrilJibril

1111

asked Apr 7 at 23:23

JibrilJibril

1111