Machine Learning Meetup Notes: 2010-04-28

From Noisebridge
(Difference between revisions)
Jump to: navigation, search
(Created page with '*Mike S presented a mathematical overview of SVMs **Started with introduction to linear classification [http://en.wikipedia.org/wiki/Linear_classifier] **Discussed the kernel tri…')
 
 
(3 intermediate revisions by one user not shown)
Line 5: Line 5:
 
**Emphasized two important aspects of SVMs:
 
**Emphasized two important aspects of SVMs:
 
***Dual problem is a quadratic programming problem that is easier to solve than the primal problem
 
***Dual problem is a quadratic programming problem that is easier to solve than the primal problem
***After dual problem is optimized, only the support vectors (the data points whose langrangian multipliers are > 0) are needed to make predictions for new data
+
***After dual problem is optimized, only the support vectors (the data points whose langrangian multipliers are > 0) are needed to make predictions for new data (and their associated multipliers)
 +
*Thomas talked about the KDD conference and their data competition [https://pslcdatashop.web.cmu.edu/KDDCup/rules_data_format.jsp]
 +
**Started a wiki page for the competition: https://www.noisebridge.net/wiki/KDD_Competition_2010
 +
*Sai skyped in and talked a bit about his use of libSVM for classification of user history on his website cssfingerprint.com
 +
*We talked a little bit about libSVM [http://www.csie.ntu.edu.tw/~cjlin/libsvm/]
 +
*Ted posted some links:
 +
**WEKA, a variety pack of ML tools written in Java: http://www.cs.waikato.ac.nz/ml/weka/
 +
**KDD Datasets: http://kdd.ics.uci.edu/summary.data.type.html

Latest revision as of 09:18, 29 April 2010

  • Mike S presented a mathematical overview of SVMs
    • Started with introduction to linear classification [1]
    • Discussed the kernel trick [2]
    • Loosely derived the loss function and dual loss function for support vector machines [3]
    • Emphasized two important aspects of SVMs:
      • Dual problem is a quadratic programming problem that is easier to solve than the primal problem
      • After dual problem is optimized, only the support vectors (the data points whose langrangian multipliers are > 0) are needed to make predictions for new data (and their associated multipliers)
  • Thomas talked about the KDD conference and their data competition [4]
  • Sai skyped in and talked a bit about his use of libSVM for classification of user history on his website cssfingerprint.com
  • We talked a little bit about libSVM [5]
  • Ted posted some links:
Personal tools