1 / 24

Matrix Factorization

Matrix Factorization. Reporter : Sun Yuanshuai E-mail : sunyuanshuai@gmail.com. 1. MF Introduction. 2. Application Area. 3. My Work. 4. Difficulty in my work. Content. MF Introduction.

curt
Télécharger la présentation

Matrix Factorization

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Matrix Factorization Reporter : Sun Yuanshuai E-mail : sunyuanshuai@gmail.com

  2. 1 MF Introduction 2 Application Area 3 My Work 4 Difficulty in my work Content

  3. MF Introduction Matrix factorization (abbr. MF), just as the name suggests, decomposes a big matrix into the multiplication form of several small matrix. It defines mathematically as follows, We here assume the target matrix , the factor matrix and , where K << min (m, n), so it is

  4. MF Introduction We quantify the quality of the approximation with the Euclidean distance, so we can get the objective function as follows, Where i.e. is the predict value.

  5. MF Introduction 1. Alternating Descent Method This method only works, when the loss function implies with Euclidean distance. So, we can get The same to .

  6. MF Introduction 2. Gradient Descent Method The update rules of U defines as follows, where The same to .

  7. MF Introduction Stochastic Gradient Algorithm Gradient Algorithm

  8. MF Introduction Online Algorithm Online-Updating Regularized Kernel Matrix FactorizationModels for Large-Scale Recommender Systems

  9. MF Introduction Loss Function We update the factor V for reducing the objective function f with the conventional gradient descendent, as follows, Here we set , so it is reachable go , the same to factor matrix U.

  10. MF Introduction Here, we go on based on an assumption that SSGD can converge to a set of stationary points.

  11. MF Introduction We see that there exists dependence between the current solution and the last one gotten by iteration operation, i.e. the last solution has to be known before the current can be computed. To solve the problem, we propose the notion interchangeablility : The idea of DSGD is to specialize the SSGD algorithm, choosing the strata with special layout such that SGD can be run on each stratum in a distributed manner.

  12. MF Introduction We can get the theorem from definition about interchangeability, as follows: From the theorem, we can compute the train matrix which is block-diagonal in parallel, i.e. can be computed independ -ently.

  13. MF Introduction We can compute the block-diagonal matrix in parallel. Our target, however, is to make the general matrix decomposition parallelism. How can we make it? Now we can stratify the input matrix, such that each stratum meets the interchangeable condition. Assume we cut input matrix into 3*3 blocks, as follows:

  14. MF Introduction

  15. Application Area Dyadic Data : In general, dyadic data are the measurements on dyads, which are pairs of two elements coming from two sets. buy = (userId, itemId, rating) product customer Any area where dyadic data can be generated.

  16. My Work

  17. My Work

  18. × = + Left Matrix × = + Right Matrix × = ||

  19. Difficulty in my work DataSet I use a total of 5 jobs. But the job can’t work. Just because the data generated in the procedure is too big which is 6000GB, Analyzed as follows, the left matrix is 300 thousand * 250 thousand approximately, the right matrix F is 250 thousand * 10 approximately, so the additional data generated is 250K*10*250K*8=6000G, where the 8 implies the number of bytes taken to store a double.

  20. Difficulty in my work • The techniques I have used: • Combiner • Compress

  21. THANK YOU FOR YOUR TIME!

More Related