Profile

Cover photo
Alexandre KOWALCZYK
Works at http://www.svm-tutorial.com
Attended Supinfo
Lives in Paris
187 followers|153,479 views
AboutPosts

Stream

Alexandre KOWALCZYK

Shared publicly  - 
 
Parts 4, 5 and 6 of my tutorial series about the math behind SVMs are online. http://bit.ly/2c3GfX9 #MachineLearning #SVM #AI
1
3
Alexandre KOWALCZYK's profile photoTariq Adnan's profile photoRishabh Chakrabarti's profile photo
8 comments
 
Hi +Alexandre KOWALCZYK. You're website svm-tutorial.com - SVM Tutorial is pretty good. Can you please tell me how I can embed and make good graphs online for a blog. 
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
Interesting
 
"Deep Learning for Computer Vision: A comparison between CNNs and Heirarchical Temporal Memories on Object recognition Tasks" - Master's thesis by Vincenzo Lomonaco http://buff.ly/1KlNxez
View original post
1
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
Check out my new article : How to classify text with SVM in R    
#SVM #R #machinelearning
http://www.svm-tutorial.com/2014/11/svm-classify-text-r
3
7
chandra sutrisno's profile photoAlexandre KOWALCZYK's profile photosonia sadiq's profile photoAkash Joshi's profile photo
6 comments
 
Sir is it possible to train a svr model using electricity data for 2 days but predict for only 1 day using R
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
1
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
The beauty of mathematics
 
A post about a paper which shows how mathematical research can deal with difficult questions using quite intuitive ideas and, on the way, producing beautiful pictures
1 comment on original post
1
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
Nice post about RNN
 
A nicely written RNN tutorial; I am really curious and looking forward to the next parts.

http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-1-introduction-to-rnns/
Recurrent Neural Networks (RNNs) are popular models that have shown great promise in many NLP tasks. But despite their recent popularity I’ve only found a limited number of resources that thr…
2 comments on original post
1
zhenhui xiang's profile photo
 
Hi I've found that you have some tutorials on SVM classification, those are good ! however, I would like  to know how to write a regression code which could help to build a  function that could represent the correlations of my data sets? do you have any sample codes on this? Thank you!
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
Excellent. 
 
“Neural Networks, Types, and Functional Programming” great overview, discussion, and just nicely written
http://colah.github.io/posts/2015-09-NN-Types-FP/
An Ad-Hoc Field. Deep learning, despite its remarkable successes, is a young field – perhaps ten years old. While models called artificial neural networks have been studied for decades, much of that work seems only tenuously connected to modern results. It's often the case that young fields ...
View original post
1
1
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
Here is the Part 2 of my tutorial about the math behind the support vector machine. We will have fun with vectors and hyperplane :)  Looking forward to hear your thoughts. 

#mathematics   #artificialintelligence   #supervisedlearning  
http://www.svm-tutorial.com/2014/11/svm-understanding-math-part-2/
In this article about SVM, after reviewing some basics about vectors, I explain how to compute the margin between a point and the hyperplane.
6
1
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
After several months here is my first post on my new blog about support vector machine. I would love to hear your thoughs. 

http://www.svm-tutorial.com/2014/10/svm-linear-kernel-good-text-classification/
A linear kernel is often recommended for text classification with SVM because text data has a lot of features and is often linearly separable.
8
Kaustav Mukherjee's profile photo
 
Man, that is the most awesome, concise video on SVM I have ever seen! Please come up with the 4th part!

Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
 
Vulnerability

   Mankind is not  as Western economists and philosophers from the 18th and 19th centuries imagined it: essentially competitive, isolated, each one related to others only by the necessity of exchanging the satisfaction of economic and instinctual needs. Pleasure is not  limited to the relief of a tension. Loving-kindness is not  an unreasonable demand, contrary to reality.
   One aspect of mankind is well described by such views, but only one aspect!

   The other aspect is that humans can take 'unreasonable' risks, love 'unknown' strangers (or even not-so-strangers but still fundamentally 'unknown'), step out of 'natural' conventions, explore…
   The other aspect is that humans crave not only for certainty (and the control and exploitation of their world to provide it), but also for expansion, for discovery, for knowledge (even when there isn't an immediate utilitarian application).
   And these moments require fragility, openness and vulnerability… These moments require trusting the world, and another: the 'Other'.
   These moments might be the most 'alive', even though they're when you less recognise 'yourself'! These moments might be the most mindful, as you monitor what's happening but without immediately liking or disliking, neither recollecting a past nor immediately projecting a future… These moments might be life out of the narratives of causality, out of 'should's, out of prejudices and preferences and habits and biases! These moments might be when you meet the Ultimate truth that you're nothing but an experience,  born out of a will  to experience.


   video: FIRST KISS (3'16'')

   This isn't about strangers without names (even a long-term partner is a 'stranger' if you don't blind yourself to impermanence!), this isn't even about relationships.
   This is about not-knowing, this is about meeting the Other (instead of keeping it at a 'safe' distance), this is about engaging with Oneness and Separateness at once;  this is about getting out of your head and being fully present to the situation at hand. This is about forgiving and trusting, stepping away from past hurt and the fear of future hurt (not by denying these but by not identifying with them, not letting them define who you embody).
   No attempt will automatically work out well, but what do you see coming if you don't even try? A 'full' life? A 'full' experience? Happiness?
   This is about being fearless (gplus.wallez.name/G28hdsxKsDP) and brave enough that you open yourself, that you accept your vulnerability, that you accept pain so you can let it go instead of fighting against it (thus carrying it around with you, doubling it by the sheer frustration of never truly becoming 'stronger' than pain).

   This is about letting the world heal you as much as you healing the world… but this starts by accepting that it is painful right now, that habits and certainties didn't create the expected results, that you don't really know what's going on but that it seems clear from the experienced results that you need to step out of the known tracks, of your own worn-out status quo…
   You can still use your head! but how much longer should you stay 'locked' in your head? Do you use mental anchors appropriately, or are the anchors impeding your movements and deciding for you how far you go?


#Buddhism  
photo: "Avenida de la Vida II", © Bertram Bahner
6 comments on original post
1
Add a comment...

Alexandre KOWALCZYK

Shared publicly  - 
 
Amazing!
 
Fascinating

Turning 3D printer into an air hockey #robot
the robot doesn’t just move in two dimensions. It can also predict the puck’s movements, decide whether to block or shoot, and is adjustable in difficulty

http://thenextweb.com/shareables/2014/02/07/guy-hacked-3d-printer-air-hockey-robot/
50 comments on original post
3
2
John Green's profile photo
 
That was Awesome!!!
Add a comment...
Story
Introduction
Hello, and welcome to my Google + Profile.  

My name is Alexandre KOWALCZYK.  I am a software engineer.

Artificial Intelligence

I am really passionate about artificial intelligence. I particularly like classifiers such as neuron networks and SVM. 


 
Bragging rights
I read all books written by Emile Zola
Education
  • Supinfo
    Computer Science, 2006 - 2011
Basic Information
Gender
Male
Looking for
Networking
Work
Skills
C#, R, SVM, Artificial Intelligence
Employment
  • http://www.svm-tutorial.com
    present
Places
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Currently
Paris
Previously
Montréal - La Ferté sous Jouarre
Links
Contributor to