手册

Chapter 01

❑人工智能:Artificial Intelligence

❑计算智能:Computational Intelligence

❑感知智能:Perceptual Intelligence

❑认知智能:Cognitive Intelligence

❑机器学习:Machine Learning

❑有监督学习:Supervised learning

❑无监督学习:Unsupervised learning

❑增强学习:Reinforcement Learning

❑ 神经元:Neuron

❑ 感知器:Perceptron

❑神经网络:Neural Networks

❑反向传播算法**: Back Propagation, BP**

❑卷积神经网络:Convolutional Neural Network，CNN

❑深度学习:Deep Learning

❑梯度消失:Vanishing Gradient

❑修正线性单元:Rectified Linear Unit, ReLU

❑深度信度网:Deep Belief Networks

❑玻尔兹曼机:Boltzmann Machines

❑变分学习:Variational Learning

❑ 分类:Classification

❑ 递归:Recursion

❑深度信念网络:Deep Belief Network, DBN

❑深度玻尔兹曼机:Deep Boltzmann Machine, DBM

❑深度自编码器**: Deep Auto-Encoder, DAE**

❑降噪自编码器**: Denoising Auto-Encoder, D-AE**

❑栈式自编码器**: Stacked Auto-Encoder, SAE**

❑生成对抗网络**: Generative Adversarial Networks**，GAN

❑非参数贝叶斯网络**: Non-parametric Bayesian Networks**

❑深度前馈网络**: Deep Feedforward Neural Network, D-FNN**

❑卷积神经网络**: Convolutional Neural Network, CNN**

❑循环神经网络**: Recurrent Neural Network, RNN**

❑胶囊网络**: Capsule Net**

❑深度森林**: Deep Forest**

❑图像分类(物体识别): Image Classification (Object Recognition)

❑物体检测:Object Detection

❑图像分割:Image Segmentation

❑图像回归:Image Regression

❑语音识别:Automatic Speech Recognition, ASR

❑声纹识别:Voiceprint Recognition

❑语音合成:Speech Synthesis

❑语言模型:Language Model

❑情感分析:Sentiment Analysis

❑神经机器翻译:Neural Machine Translation, NMT

❑神经自动摘要:Neural Automatic Summarization

❑机器阅读理解:Machine Reading Comprehension, MRC

❑自然语言推理:Natural Language Inference, NLI

❑文本蕴含:Text Entailment

Chapter 02

❑ 标量:Scalar

❑向量:Vector

❑张量:Tensor

❑ 期望:Expectation

❑方差:Variance

❑ 熵:Entropy

❑ 过拟合:Overfitting

❑ 欠拟合:Underfitting

❑监督学习:Supervised learning

❑无监督学习:Unsupervised learning

❑数据集:Data set

❑训练集:Training set

❑验证集:Validation set

❑测试集:Testing set

❑ 泛化:Generalization

❑线性回归:Linear Regression

❑支持向量机:Support Vector Machine

❑决策树:Decision Tree

❑随机森林: Random Forest

❑ 感知器:Perceptron

❑反向传播算法**: (error) Back Propagation, BP**

❑梯度下降:Gradient Descent

❑修正线性单元:Rectified Linear Unit, ReLU

Chapter 03

❑ 卷积:Convolution

❑ 神经认知机:Neocognitron

❑ 感受野:Receptive field

❑ 感光细胞:Photoreceptor cell

❑ 水平细胞:Horizontal cell

❑ 双极细胞:Bipolar cell

❑ 神经节细胞:Ganglion cell

❑ 示波器:Oscilloscope

❑ 电极:Electrode

❑ 视网膜:Retina

❑ 视觉皮层:Visual cortex

❑ 视神经:Optic nerve

❑ 外侧膝状体:Lateral Geniculate Nucleus

❑ 卷积神经网络:Convolutional Neural Network

❑ 卷积核:Convolutional kernel

❑ 池化:Pooling

❑ 池化核:Pooling kernel

❑ 零填充:Zero-padding

❑ 特征图:Feature map

❑ 步幅:Stride

❑ 降采样:Down sampling

❑ 最大池化:Max-pooling

❑ 均值池化:Average-pooling

❑ 残差神经网络:Residual Neural Network

❑ 残差块:Residual block

❑ 跳跃连接:Skip connection

❑ 径向基函数:Radial Basis Function，RBF

❑ 区域CNN: Region-CNN

❑ 选择性搜索:Selective Search

❑ 区域建议网络:Region Proposal Network，RPN

❑ 边框回归:Bounding Box Regression

Chapter 04

❑ 计算图:Computational graph

❑ 循环神经网络:Recurrent Neural Network

❑ 随时间反向传播算法:BP Through Time, BPTT

❑ 长短时记忆网络:Long Short-Term Memory

❑ 遗忘门:Forget gate

❑ 输入门:Input gate

❑ 输出门:Output gate

❑ 双向RNN:Bidirectional RNN

❑ 门控循环单元:Gated Recurrent Unit (GRU)

❑ 窥孔LSTM: Peephole LSTM

❑ 连续时间RNN: Continuous time RNN

❑ 语言模型:Language model

❑ 神经机器翻译**: Neural Machine Translation**

❑ 图像描述:Image captioning

❑ 自动摘要:Automatic summarization

❑ 自动写作:Automatic writing