User Tools

Site Tools


Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
papers:bottou-91a [2007/08/16 14:43]
leonb
papers:bottou-91a [2020/01/09 10:41] (current)
leonb
Line 7: Line 7:
 relatively new ideas for the time. relatively new ideas for the time.
  
-   * [[wp>Connectionism|Connectionist learning algorithms]] can be studied as [[wp>stochastic approximations]] and more specifically [[wp>stochastic gradient descent]] algorithms, possibly involving surrogate loss functions (chapters 2 and 3.+   * [[wp>Connectionism|Connectionist learning algorithms]] can be studied as [[wp>stochastic approximations]] and more specifically [[wp>stochastic gradient descent]] algorithms, possibly involving surrogate loss functions (chapters 2 and 3). 
-   * The performance of these algorithms can be studied using advanced statistical theories such as [[wp>Vladimir Vapnik]]'s [[wp>Structural Risk Minimization]] (chapter 4.)+   * The performance of these algorithms can be studied using advanced statistical theories such as [[wp>Vladimir Vapnik]]'s [[wp>Structural Risk Minimization]] (chapter 4).
    * The speed of these algorithms can be approached using the methods of numerical optimization (chapter 5.)    * The speed of these algorithms can be approached using the methods of numerical optimization (chapter 5.)
-   * Complex applications, such as [[wp>speech recognition]], can be addressed using modular learning systems such as Discriminant Hidden Markov Models (chapter 8 and 9.+   * Complex applications, such as [[wp>speech recognition]], can be addressed using modular learning systems such as Discriminant Hidden Markov Models (chapter 8 and 9). 
-   * Probabilistic discriminant modular learning systems are often limited by the so-called //label-bias problem// (chapter 10). It took a few more years [[:research:structured|to find a solution.]]+   * Probabilistic discriminant modular learning systems are often limited by the so-called //label-bias problem// (chapter 10).  Finding a solution took a [[:research:structured|three more years]]
 +==== ====
  
  
 <box 99% orange> <box 99% orange>
-Léon Bottou:  //**Une Approche théorique de l'Apprentissage Connexionniste: Applications à la Reconnaissance de la Parole**//, Orsay, France, 1991.+Léon Bottou:  //**Une Approche théorique de l'Apprentissage Connexionniste: Applications à la Reconnaissance de la Parole**//, Ph.D. thesis, Université de Paris XI, Orsay, France, 1991.
  
-[[http://leon.bottou.org/publications/djvu/bottou-1991.djvu|bottou-1991.djvu]]+[[http://leon.bottou.org/publications/djvu/bottou-1991.djvu|bottou-1991.djvu (one big file)]]
 [[http://leon.bottou.org/publications/pdf/bottou-1991.pdf|bottou-1991.pdf]] [[http://leon.bottou.org/publications/pdf/bottou-1991.pdf|bottou-1991.pdf]]
-[[http://leon.bottou.org/publications/psgz/bottou-1991.ps.gz|bottou-1991.ps.gz]]+[[http://leon.bottou.org/publications/psgz/bottou-1991.ps.gz|bottou-1991.ps.gz]]\\ 
 +[[http://leon.bottou.org/cgi/djvuserve/publications/djvu/bottou-1991.djvu|bottou-1991.djvu (served page per page)]]
 </box> </box>
  
papers/bottou-91a.1187289794.txt.gz · Last modified: 2007/08/16 14:43 by leonb

Page Tools