« IML » : différence entre les versions

De Parcours SIIA
Aller à la navigation Aller à la recherche
mAucun résumé des modifications
 
(71 versions intermédiaires par 5 utilisateurs non affichées)
Ligne 1 : Ligne 1 :
IML : Interactive Machine Learning
== '''IML : Interactive Machine Learning''' ==


=== Coordination : ===
Cédric Buche (buche@enib.fr)
Cédric Buche (buche@enib.fr)


Lesson 1 :  
=== Presentation : ===
[[Fichier:IML Introduction lesson1.pdf|vignette]]
IML merges machine learning and human-computer interaction. While traditional machine learning systems process the data that have been given to them in advance, this course considers that the learning process could benefit from interactions with the environment as well as with a human, and that inputs and outputs from and for humans carry meaningful information. Indeed humans may provide input to a learning algorithm, including inputs in the form of labels, demonstrations, advice, rewards or rankings. The interaction is all the more useful as the human can guide along the learning process while adapting his guidance to the outputs of the algorithm. This interaction can be in the form of feedforward or feedback information. The timing of these interactions can be preset, left to the teacher’s initiative or even to the learner’s initiative. In the latter case, the algorithm called “active learner" can decide when,about what, how and with whom to interact to optimise its learning process. Thus a bidirectional dialogue can emerge. Another point addressed by this course is the self-representation of the machine through its interaction with its environment. According to cognitive science, our skills and our representations of ‘a’ world come from our sensorimotor capabilities. These latter are the basis to a more abstract representation and some researches aims to propose models and algorithms that reproduce the underlying mechanisms. Indeed, these mechanisms could provide a mean to resolve the problem of the knowledge representations of unknown environment. The models allows for an anticipation of the perception from the action of the agent and for a construction of its own world. These researches are generally inspired by cognitive science.


Lesson 2 :  
=== Content : ===
[[Fichier:IML clustering classif feature test-min.pdf|vignette]]


Lesson 3 :
<!--
[[Fichier:IML navigation.pdf|vignette]]


Lesson 4 :  
'''Schedule''' : [[Fichier:Siia_iml_schedule-2020.pdf|200px|thumb|left|texte descriptif]]
[[Fichier:Cours_metaHeuristique_cb_print.pdf|vignette]]


Lesson 5:
Principles of interactive learning: imitation learning and active requests


Lesson 6:
'''Lesson 0''' (09/15) : [[Fichier:Cours_welcome_IML.pdf|200px|thumb|left|texte descriptif]]
[[Cognitive developmental approaches for sociable artificial agents]]
 
 
'''Lesson 1''' (09/15) : [[Fichier:Cours_introduction_IML.pdf|200px|thumb|left|texte descriptif]]
 
 
Lesson 1 - code (09/15) : [[Fichier:Code_lesson1.zip|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 2''' : [[Fichier:Cours_data_IML_cb.pdf|200px|thumb|left|texte descriptif]]
 
Lesson 2 - code : [[Fichier:Code_cours2.zip|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 3''' : [[Fichier:Cours_test_feature_IML_cb.pdf|200px|thumb|left|texte descriptif]]
 
Lesson 3 - code : [[Fichier:Code_3.zip|200px|thumb|left|texte descriptif]]
 
 
'''Lab  1'''  : [[Fichier:Iml_galaxie_tp.pdf|200px|thumb|left|texte descriptif]]
 
Lab  1 - correction  : [[Fichier:CorrectionGalaxie.zip|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 4''' : [[Fichier:HRI.pdf|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 5''' : [[Fichier:ApprentissageComportement_1.pdf|200px|thumb|left|texte descriptif]]
 
 
'''Lab  2''' (05/11)  : [[Fichier:iml_robot_tp2.pdf|200px|thumb|left|texte descriptif]]
 
Lab  2 - code  : [[Fichier:urdf.zip|200px|thumb|left|texte descriptif]]
 
 
'''PROJECT ''' : You need to complete a project using qiBullet. The two expected components are
*  human / robot interaction
*  machine learning
. The result will be (12/03/2020) : 
*  a demonstration during the defense 
*  a video 
*  source code
 
=== Grade : ===
https://docs.google.com/spreadsheets/d/1DJlV6VNF88-pHqjiVF9LI0zLtgYDnnK7cD6JQyJuQYE/edit?usp=sharing
 
-->
 
 
=== Grade : ===
 
[[Fichier:Notes_IML_2019_2.pdf|200px|thumb|left|texte descriptif]]
 
=== IML 2020-2021 ===
 
=== Content : ===
 
'''Schedule''' : [[Fichier:Siia_iml_schedule-2020.pdf|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 0''' (09/15) : [[Fichier:Cours_welcome_IML.pdf|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 1''' (09/15) : [[Fichier:Cours_introduction_IML.pdf|200px|thumb|left|texte descriptif]]
 
'''Version modifiée par Mr De Loor''' (09/15) : [[Fichier:Cours_introduction_IMLPDL.pdf|200px|thumb|left|texte descriptif]]
 
 
Lesson 1 - code (09/15) : [[Fichier:Code_lesson1.zip|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 2''' : [[Fichier:Cours_data_IML_cb.pdf|200px|thumb|left|texte descriptif]]
 
Lesson 2 - code : [[Fichier:Code_cours2.zip|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 3''' : [[Fichier:Cours_test_feature_IML_cb.pdf|200px|thumb|left|texte descriptif]]
 
Lesson 3 - code : [[Fichier:Code_3.zip|200px|thumb|left|texte descriptif]]
 
 
'''Lab  1'''  : [[Fichier:Iml_galaxie_tp.pdf|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 4''' : [[Fichier:HRI.pdf|200px|thumb|left|texte descriptif]]
 
 
'''Lesson 5''' : [[Fichier:ApprentissageComportement_1.pdf|200px|thumb|left|texte descriptif]]
 
 
'''Lab  2''' (05/11)  : [[Fichier:iml_robot_tp2.pdf|200px|thumb|left|texte descriptif]]
 
Lab  2 - code  : [[Fichier:urdf.zip|200px|thumb|left|texte descriptif]]
 
 
'''PROJECT ''' : You need to complete a project using qiBullet. The two expected components are
*  human / robot interaction
*  machine learning
. The result will be (12/03/2020) : 
*  a demonstration during the defense 
*  a video 
*  source code
 
=== Grade : ===
https://docs.google.com/spreadsheets/d/1DJlV6VNF88-pHqjiVF9LI0zLtgYDnnK7cD6JQyJuQYE/edit?usp=sharing

Dernière version du 22 septembre 2021 à 10:21

IML : Interactive Machine Learning

Coordination :

Cédric Buche (buche@enib.fr)

Presentation :

IML merges machine learning and human-computer interaction. While traditional machine learning systems process the data that have been given to them in advance, this course considers that the learning process could benefit from interactions with the environment as well as with a human, and that inputs and outputs from and for humans carry meaningful information. Indeed humans may provide input to a learning algorithm, including inputs in the form of labels, demonstrations, advice, rewards or rankings. The interaction is all the more useful as the human can guide along the learning process while adapting his guidance to the outputs of the algorithm. This interaction can be in the form of feedforward or feedback information. The timing of these interactions can be preset, left to the teacher’s initiative or even to the learner’s initiative. In the latter case, the algorithm called “active learner" can decide when,about what, how and with whom to interact to optimise its learning process. Thus a bidirectional dialogue can emerge. Another point addressed by this course is the self-representation of the machine through its interaction with its environment. According to cognitive science, our skills and our representations of ‘a’ world come from our sensorimotor capabilities. These latter are the basis to a more abstract representation and some researches aims to propose models and algorithms that reproduce the underlying mechanisms. Indeed, these mechanisms could provide a mean to resolve the problem of the knowledge representations of unknown environment. The models allows for an anticipation of the perception from the action of the agent and for a construction of its own world. These researches are generally inspired by cognitive science.

Content :

Grade :

Fichier:Notes IML 2019 2.pdf

IML 2020-2021

Content :

Schedule : Fichier:Siia iml schedule-2020.pdf


Lesson 0 (09/15) : Fichier:Cours welcome IML.pdf


Lesson 1 (09/15) : Fichier:Cours introduction IML.pdf

Version modifiée par Mr De Loor (09/15) : Fichier:Cours introduction IMLPDL.pdf


Lesson 1 - code (09/15) : Fichier:Code lesson1.zip


Lesson 2 : Fichier:Cours data IML cb.pdf

Lesson 2 - code : Fichier:Code cours2.zip


Lesson 3 : Fichier:Cours test feature IML cb.pdf

Lesson 3 - code : Fichier:Code 3.zip


Lab 1  : Fichier:Iml galaxie tp.pdf


Lesson 4 : Fichier:HRI.pdf


Lesson 5 : Fichier:ApprentissageComportement 1.pdf


Lab 2 (05/11)  : Fichier:Iml robot tp2.pdf

Lab 2 - code  : Fichier:Urdf.zip


PROJECT  : You need to complete a project using qiBullet. The two expected components are

  • human / robot interaction
  • machine learning

. The result will be (12/03/2020) :

  • a demonstration during the defense
  • a video
  • source code

Grade :

https://docs.google.com/spreadsheets/d/1DJlV6VNF88-pHqjiVF9LI0zLtgYDnnK7cD6JQyJuQYE/edit?usp=sharing

Récupérée de « http://siia.univ-brest.fr/w/index.php?title=IML&oldid=973 »