人工智能算法·基本算法■神经网络■深度学习■原则上各种机器学习算法都可以使用·拓展算法·迁移学习·增强学习，对抗学习·小样本学习·因果推断
▪ 基本算法 ▪ 神经网络 ▪ 深度学习 ▪ 原则上各种机器学习算法都可以使用 ▪ 拓展算法 ▪ 迁移学习 ▪ 增强学习 ▪ 对抗学习 ▪ 小样本学习 ▪ 因果推断 人工智能算法
Neural Network (NN)节点，连接How does neuralnetworkwork?XStandard NN干亿神经元，每个约1000个连接Mimichumanneuralsystem
▪ How does neural network work? 节点，连接 ▪ Mimic human neural system 千亿神经元，每个约1000个连接 Neural Network (NN)
Artificial Neural Networks (ANN)Black boxX1YX2X3Input0001X,-0111Output11011111X2-V000100100111X30000Output Y is 1 if at least two of the three inputs are equal to 1
Artificial Neural Networks (ANN) X1 X2 X3 Y 1 0 0 0 1 0 1 1 1 1 0 1 1 1 1 1 0 0 1 0 0 1 0 0 0 1 1 1 0 0 0 0 X1 X2 X3 Black box Output Y Input Output Y is 1 if at least two of the three inputs are equal to 1
Artificial Neural Networks (ANNInputnodesBlack boxX1X3X2YOutput00100.301node1110110.31111￥2Y001001000111X30.3t=0.40000Y = I(0.3X, + 0.3X2 +0.3X3 - 0.4 > 0)if z istrue1where I(z)0otherwise
Artificial Neural Networks (ANN) X1 X2 X3 Y 1 0 0 0 1 0 1 1 1 1 0 1 1 1 1 1 0 0 1 0 0 1 0 0 0 1 1 1 0 0 0 0 X1 X2 X3 Y Black box 0.3 0.3 0.3 t=0.4 Output node Input nodes − 0.4 0) if z istrue 0 otherwise where I (z) = 1 Y = I (0.3X1 + 0.3X 2 + 0.3X 3
Artificial Neural Networks (ANN)InputnodesBlack box: Model is an assembly ofOutputXinter-connected nodesnodeW1W2and weighted linksXYXW3- Output node sums up专+each of its input valueaccording to the weightsPerceptron Modelof its linksY =I(Zw,X,-t)or: Compare output nodeagainst some threshold t Y = sign(w,X, -t)
Artificial Neural Networks (ANN) ▪ Model is an assembly of inter-connected nodes and weighted links ▪ Output node sums up each of its input value according to the weights of its links ▪ Compare output node X1 X2 X3 Y Black box w1 t Output node Input nodes w2 w3 i Y = i i I(w X −t) PerceptronModel i − t) against some threshold t Y = sign(wiXi or