教學大綱表
請遵守智慧財產權,勿使用非法影印教科書,避免觸法。
課程名稱 (中文) 類神經網路概論
(英文) Introduction To Neural Networks
開課單位 資訊工程學系
課程代碼 I4850
授課教師 虞台文
學分數 3.0 必/選修 選修 開課年級 大四
先修科目或先備能力:Linear Algebra, Probability
課程概述與目標: This course provides an introduction to concepts in neural networks and connectionist models. Topics include parallel distributed processing, learning algorithms, and applications. Specific networks discussed include Hopfield networks, bidirectional associative memories, perceptrons, feedforward networks with back propagation, and competitive learning networks, including self-organizing and Grossberg networks. Software for some networks is provided.
教科書
課程大綱 學生學習目標 單元學習活動 學習成效評量 備註
單元主題 內容綱要
1 Introduction to neural networks 1. Objective of neural networks
2. Neural network models
3. Neural networks vs. pattern classification
4. Neural network vs. machine learning
1. Objective of neural networks
2. Neural network models
3. Neural networks vs. pattern classification
4. Neural network vs. machine learning
 
2 Feedforward Neural Networks 1. Artificial neuron models
2. Single-layer perceptron networks
3. learning rules for single-layer perceptron networks
- perceptron learning rule
- adaline leaning rule
- delta-leaning rule
1. Artificial neuron models
2. Single-layer perceptron networks
3. learning rules for single-layer perceptron networks
- perceptron learning rule
- adaline leaning rule
- delta-leaning rule
 
3 Multilayer Perceptron 1. The structure of multilayer perceptron
2. XOR problem
3. Non-inear separability of MLP
4. Back propagation learning algorithm
1. The structure of multilayer perceptron
2. XOR problem
3. Non-inear separability of MLP
4. Back propagation learning algorithm
 
4 Multilayer Perceptron 1. The structure of multilayer perceptron
2. XOR problem
3. Non-inear separability of MLP
4. Back propagation learning algorithm
1. The structure of multilayer perceptron
2. XOR problem
3. Non-inear separability of MLP
4. Back propagation learning algorithm
 
5 Radial basis function network 1. Network architecture
2. Training
3. Applicaitons
- Logistic map
- Function approximation
- Time series prediction
- Control of a chaotic time series
1. Network architecture
2. Training
3. Applicaitons
- Logistic map
- Function approximation
- Time series prediction
- Control of a chaotic time series
 
6 Unsupervised Learning 1. Clustering
2. K-means algorithm
3. Competitive learning
4. Kohonon self-organizing nets
5. Principal component analysis (PCA)
1. Clustering
2. K-means algorithm
3. Competitive learning
4. Kohonon self-organizing nets
5. Principal component analysis (PCA)
 
7 Unsupervised Learning 1. Clustering
2. K-means algorithm
3. Competitive learning
4. Kohonon self-organizing nets
5. Principal component analysis (PCA)
1. Clustering
2. K-means algorithm
3. Competitive learning
4. Kohonon self-organizing nets
5. Principal component analysis (PCA)
 
8 Fuzzy Logic 1. Introdunction to fuzzy logic
1. Introduction to fuzzy logic
2. Operations on fuzzy sets
3. Fuzzy relations
- The extension principle
- Metrics for fuzzy numbers
- Fuzzy implications
- Linguistic variables
1. Introdunction to fuzzy logic
1. Introduction to fuzzy logic
2. Operations on fuzzy sets
3. Fuzzy relations
- The extension principle
- Metrics for fuzzy numbers
- Fuzzy implications
- Linguistic variables
 
9 Fuzzy Logic 1. Introdunction to fuzzy logic
1. Introduction to fuzzy logic
2. Operations on fuzzy sets
3. Fuzzy relations
- The extension principle
- Metrics for fuzzy numbers
- Fuzzy implications
- Linguistic variables
1. Introdunction to fuzzy logic
1. Introduction to fuzzy logic
2. Operations on fuzzy sets
3. Fuzzy relations
- The extension principle
- Metrics for fuzzy numbers
- Fuzzy implications
- Linguistic variables
 
10 Fuzzy Neural Networks 1. Integration of fuzzy logic and neural networks
2. Fuzzy neurons
3. Hybrid neural nets
4. Fuzzy neural networks as universal approximators
1. Integration of fuzzy logic and neural networks
2. Fuzzy neurons
3. Hybrid neural nets
4. Fuzzy neural networks as universal approximators
 
11 Convolution neural networks 1. Invariant pattern recognition
2. Neocognitron
3. Convolution neural networks for face recognition
1. Invariant pattern recognition
2. Neocognitron
3. Convolution neural networks for face recognition
 
12 Feeback Neural networks 1. Recurrent NNs
2. Architectures
- Elman network
- Jordan network
- Hopfield network
- Recurrent Multilayer Perceptron
3. Training
- Backpropagation through time
- Real-time recurrent learning
- Genetic algorithms
1. Recurrent NNs
2. Architectures
- Elman network
- Jordan network
- Hopfield network
- Recurrent Multilayer Perceptron
3. Training
- Backpropagation through time
- Real-time recurrent learning
- Genetic algorithms
 
13 Feeback Neural networks 1. Recurrent NNs
2. Architectures
- Elman network
- Jordan network
- Hopfield network
- Recurrent Multilayer Perceptron
3. Training
- Backpropagation through time
- Real-time recurrent learning
- Genetic algorithms
1. Recurrent NNs
2. Architectures
- Elman network
- Jordan network
- Hopfield network
- Recurrent Multilayer Perceptron
3. Training
- Backpropagation through time
- Real-time recurrent learning
- Genetic algorithms
 
14 Hopfield Neural Network 1. Associative Memory
2. Storage algorithom for Hopfield NN
3. Hopfield NN for optimization
4. Local-minima in Hopfield NN
5. Simulated annealing
6. Boltzmann machines
1. Associative Memory
2. Storage algorithom for Hopfield NN
3. Hopfield NN for optimization
4. Local-minima in Hopfield NN
5. Simulated annealing
6. Boltzmann machines
 
15 Hopfield Neural Network 1. Associative Memory
2. Storage algorithom for Hopfield NN
3. Hopfield NN for optimization
4. Local-minima in Hopfield NN
5. Simulated annealing
6. Boltzmann machines
1. Associative Memory
2. Storage algorithom for Hopfield NN
3. Hopfield NN for optimization
4. Local-minima in Hopfield NN
5. Simulated annealing
6. Boltzmann machines
 
16 Quantum Neural Networks 1. Introduction to Quantum neuron (Q'tron)
2. Noise-injection mechanism
3. Complete theorem
1. Introduction to Quantum neuron (Q'tron)
2. Noise-injection mechanism
3. Complete theorem
 
17 Applications of Q'tron Neural Networks 1. Transportation problems
2. Knapsack problems
3. Visual cryptography
4. Constraint satisfaction
- n-Queen problem
- Sudoku solver
1. Transportation problems
2. Knapsack problems
3. Visual cryptography
4. Constraint satisfaction
- n-Queen problem
- Sudoku solver
 
18 Applications of Q'tron Neural Networks 1. Transportation problems
2. Knapsack problems
3. Visual cryptography
4. Constraint satisfaction
- n-Queen problem
- Sudoku solver
1. Transportation problems
2. Knapsack problems
3. Visual cryptography
4. Constraint satisfaction
- n-Queen problem
- Sudoku solver
 

教學要點概述:
教材編選: ■ 自編教材 □ 教科書作者提供
評量方法: 期末考:30%   期中考:30%   平時考:40%  
教學資源: ■ 教材電子檔 ■ 課程網站
扣考規定:http://eboard.ttu.edu.tw/ttuwebpost/showcontent-news.php?id=504

大學部
核心能力 期末考 期中考 平時考
核心能力一 具備運用數學、科學及資訊工程相關知識的能力。 3/10 3 3 3
核心能力二 具備設計與執行實驗,及分析與解釋數據的能力。 3/10 3 3 3
核心能力三 具備工程實務流程規劃與資訊軟硬體系統整合的能力。 2/10 2 2 2
核心能力五 具備適應職場變化的能力與持續終身學習的習慣。 2/10 2 2 2