Differences

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

--- ai:methods [2007/05/16 19:40]
cyril
+++ — (current)
@@ Line 1: / Line 1: @@
-====== Main methods in AI and robotics ======
-This is a classification of techniques and algorithms, giving only keywords ...
-===== Learning =====
-==== Supervised learning ====
-=== Classification ===
-  * **MLP (Multi Layers Perceptron)** - //PMC (Perceptron multicouches)//
-    * gradient backpropagation - //rétropropagation du gradient//
-      * stochastic
-      * with inertia
-      * simulated annealing - //recuit simulé//
-    * newton (second order)
-  * **RBFNN (Radial Basis Functions Neural Networks)**
-    * k-means then gradient descent
-    * incremental addition of neurons then exact method
-  * **SVM (Support Vectors Machine)**
-  * **Decision tree** - //arbre de décision//
-    * ID3 (based on entropy)
-  * **k-nearest neighbors** - //k plus proches voisins//
-=== Regression ===
-  * **MLP (Multi Layers Perceptron)**
-  * **RBFNN (Radial Basis Functions Neural Network)**
-  * **SVR (Support Vectors Regressor)**
-==== Unsupervised learning ====
-=== Vector quantization / Clustering ===
-[[http://www.comp.lancs.ac.uk/~kristof/research/notes/clustr/index.html|Methods Overview [EN] ]]
-  * **Sequential leader**
-  * **k-means** - //k-moyennes//
-  * **GNG (Growing Neural Gas)**
-  * **Auto-organizing maps (Kohonen)** - //cartes auto-organisatrices de Kohonen//
-==== Reinforcement learning ====
-[[http://www.cs.ualberta.ca/~sutton/book/ebook/the-book.html|Introductory Book [EN] ]]
-  * **MDP (Markov Decision Processes)**
-    * Q-learning
-    * Value iteration
-    * Policy iteration
-===== Planification =====
-==== Symbolic ====
-=== State space search ===
-    * A*, WA*, IDA*
-    * Dijkstra
-=== Logics ===
-Based on STRIPS-like languages (ADL, PDDL).
-  * **GraphPlan**, **Stan**, **IPP**, **SGP**
-  * **SATplan (Satisfiability Planning)**
-==== Others ====
-  * **Genetic algorithms** - //algorithmes génétiques//
-  * **Ant colony** - //colonies de fourmis//
-==== Specific ====
-=== Path planning ===
-  * **Configurations space**
-  * **Potential fields**
-===== Perception =====
-==== Vision ====
-=== Color Quantization ===
-  * **RGB cone**, **YUV polygon**, **HSV rectangle** : [[http://www.cs.cmu.edu/~trb/papers/wirevision00.pdf|Fast Impl. [EN] ]]
-=== Image segmentation ===
-  * **Floodfill** : [[http://www.cs.cmu.edu/~trb/papers/wirevision00.pdf|Fast Impl. [EN] ]]
-  * **Watershed** - //lignes de partage des eaux// : [[http://cmm.ensmp.fr/~beucher/wtshed.html|Method [EN] ]] [[http://www.caip.rutgers.edu/~comanici/segm_images.html|Robust Impl. [EN] ]]
-=== Filters ===
-[[http://www.ph.tn.tudelft.nl/Courses/FIP/noframes/fip.html|Image Processing Fundamentals [EN] ]]
-  * anti-noise (smoothing) : Vector Median Filter, Kuwahara filter, Peer Group Filtering, Anisotropic Filtering    [[http://www.csse.monash.edu.au/hons/projects/2002/Alexander.Wolf/pictures.html|Definitions [EN] ]] [[http://www.ph.tn.tudelft.nl/Courses/FIP/noframes/fip-Smoothin.html|Definitions [EN] ]] [[http://agni.ece.ucsb.edu/publications/99ISCAS.pdf|PGF [EN] ]]
-  * gradient : prewitt, sobel, laplace, scharr
-  * morphological : dilation, erosion, opening, closing [[http://www.ph.tn.tudelft.nl/Courses/FIP/noframes/fip-Morpholo.html|Definitions [EN] ]]
-=== Edge Detection ===
-  * Canny
-    * Canny-Deriche
-==== Sensors fusion ====
-  * **Kalman filter**
-  * **Particles filter** (bayesian network) - //filtrage particulaire//