X
机器学习 吴恩达
共113课时 19小时28分58秒秒
简介
此课程将广泛介绍机器学习、数据挖掘与统计模式识别的知识。主题包括：(i) 监督学习（参数/非参数算法、支持向量机、内核、神经网络）。(ii) 非监督学习（聚类、降维、推荐系统、深度学习）。(iii) 机器学习的优秀案例（偏差/方差理论；机器学习和人工智能的创新过程）课程将拮取案例研究与应用，学习如何将学习算法应用到智能机器人（观感，控制）、文字理解（网页搜索，防垃圾邮件）、计算机视觉、医学信息学、音频、数据挖掘及其他领域上。
吴恩达 Andrew Ng, Co-founder, Coursera; Adjunct Professor, Stanford University; formerly head of Baidu AI Group/Google Brain
章节
- 课时1：Welcome (6分54秒)
- 课时2：What is Machine Learning (7分14秒)
- 课时3：Supervised Learning (12分29秒)
- 课时4：Unsupervised Learning (14分13秒)
- 课时5：Model Representation (8分10秒)
- 课时6：Cost Function (8分12秒)
- 课时7：Cost Function - Intuition I (11分9秒)
- 课时8：Cost Function - Intuition II (8分48秒)
- 课时9：Gradient Descent (11分30秒)
- 课时10：Gradient Descent Intuition (11分50秒)
- 课时11：Gradient Descent For Linear Regression (10分20秒)
- 课时12：What-'s Next (5分49秒)
- 课时13：Matrices and Vectors (8分45秒)
- 课时14：Addition and Scalar Multiplication (6分53秒)
- 课时15：Matrix Vector Multiplication (13分39秒)
- 课时16：Matrix Matrix Multiplication (11分9秒)
- 课时17：Matrix Multiplication Properties (9分2秒)
- 课时18：Inverse and Transpose (11分12秒)
- 课时19：Multiple Features (8分22秒)
- 课时20：Gradient Descent for Multiple Variables (5分4秒)
- 课时21：Gradient Descent in Practice I - Feature Scaling (8分51秒)
- 课时22：Gradient Descent in Practice II - Learning Rate (8分58秒)
- 课时23：Features and Polynomial Regression (7分39秒)
- 课时24：Normal Equation (16分17秒)
- 课时25：Normal Equation Noninvertibility (Optional) (5分58秒)
- 课时26：Basic Operations (13分59秒)
- 课时27：Moving Data Around (16分7秒)
- 课时28：Computing on Data (13分14秒)
- 课时29：Plotting Data (9分38秒)
- 课时30：Control Statements- for, while, if statements (12分55秒)
- 课时31：Vectorization (13分48秒)
- 课时32：Working on and Submitting Programming Exercises (3分33秒)
- 课时33：Classification (8分8秒)
- 课时34：Hypothesis Representation (7分24秒)
- 课时35：Decision Boundary (14分49秒)
- 课时36：Cost Function (11分25秒)
- 课时37：Simplified Cost Function and Gradient Descent (10分14秒)
- 课时38：Advanced Optimization (14分6秒)
- 课时39：Multiclass Classification- One-vs-all (6分15秒)
- 课时40：The Problem of Overfitting (9分42秒)
- 课时41：Cost Function (10分10秒)
- 课时42：Regularized Linear Regression (10分40秒)
- 课时43：Regularized Logistic Regression (8分33秒)
- 课时44：Non-linear Hypotheses (9分35秒)
- 课时45：Neurons and the Brain (7分47秒)
- 课时46：Model Representation I (12分1秒)
- 课时47：Model Representation II (11分46秒)
- 课时48：Examples and Intuitions I (7分15秒)
- 课时49：Examples and Intuitions II (10分20秒)
- 课时50：Multiclass Classification (3分51秒)
- 课时51：Cost Function (6分43秒)
- 课时52：Backpropagation Algorithm (11分59秒)
- 课时53：Backpropagation Intuition (12分44秒)
- 课时54：Implementation Note- Unrolling Parameters (7分47秒)
- 课时55：Gradient Checking (11分37秒)
- 课时56：Random Initialization (6分51秒)
- 课时57：Putting It Together (13分23秒)
- 课时58：Autonomous Driving (6分30秒)
- 课时59：Deciding What to Try Next (5分50秒)
- 课时60：Evaluating a Hypothesis (7分35秒)
- 课时61：Model Selection and Train-Validation-Test Sets (12分3秒)
- 课时62：Diagnosing Bias vs. Variance (7分42秒)
- 课时63：Regularization and Bias-Variance (11分20秒)
- 课时64：Learning Curves (11分53秒)
- 课时65：Deciding What to Do Next Revisited (6分50秒)
- 课时66：Prioritizing What to Work On (9分29秒)
- 课时67：Error Analysis (13分11秒)
- 课时68：Error Metrics for Skewed Classes (11分35秒)
- 课时69：Trading Off Precision and Recall (14分5秒)
- 课时70：Data For Machine Learning (11分9秒)
- 课时71：Optimization Objective (14分47秒)
- 课时72：Large Margin Intuition (10分36秒)
- 课时73：Mathematics Behind Large Margin Classification (Optional) (19分41秒)
- 课时74：Kernels I (15分44秒)
- 课时75：Kernels II (15分43秒)
- 课时76：Using An SVM (21分2秒)
- 课时77：Unsupervised Learning- Introduction (3分16秒)
- 课时78：K-Means Algorithm (12分32秒)
- 课时79：Optimization Objective (7分4秒)
- 课时80：Random Initialization (7分49秒)
- 课时81：Choosing the Number of Clusters (8分22秒)
- 课时82：Motivation I- Data Compression (10分9秒)
- 课时83：Motivation II- Visualization (5分27秒)
- 课时84：Principal Component Analysis Problem Formulation (9分5秒)
- 课时85：Principal Component Analysis Algorithm (15分13秒)
- 课时86：Choosing the Number of Principal Components (10分30秒)
- 课时87：Reconstruction from Compressed Representation (3分54秒)
- 课时88：Advice for Applying PCA (12分48秒)
- 课时89：Problem Motivation (7分38秒)
- 课时90：Gaussian Distribution (10分27秒)
- 课时91：Algorithm (12分2秒)
- 课时92：Developing and Evaluating an Anomaly Detection System (13分7秒)
- 课时93：Anomaly Detection vs. Supervised Learning (7分36秒)
- 课时94：Choosing What Features to Use (12分17秒)
- 课时95：Multivariate Gaussian Distribution (Optional) (13分45秒)
- 课时96：Anomaly Detection using the Multivariate Gaussian Distribution (Optional) (14分3秒)
- 课时97：Problem Formulation (7分54秒)
- 课时98：Content Based Recommendations (14分31秒)
- 课时99：Collaborative Filtering (10分14秒)
- 课时100：Collaborative Filtering Algorithm (8分26秒)
- 课时101：Vectorization- Low Rank Matrix Factorization (8分27秒)
- 课时102：Implementational Detail- Mean Normalization (8分30秒)
- 课时103：Learning With Large Datasets (5分44秒)
- 课时104：Stochastic Gradient Descent (13分19秒)
- 课时105：Mini-Batch Gradient Descent (6分17秒)
- 课时106：Stochastic Gradient Descent Convergence (11分31秒)
- 课时107：Online Learning (12分50秒)
- 课时108：Map Reduce and Data Parallelism (14分8秒)
- 课时109：Problem Description and Pipeline (7分2秒)
- 课时110：Sliding Windows (14分39秒)
- 课时111：Getting Lots of Data and Artificial Data (16分20秒)
- 课时112：Ceiling Analysis- What Part of the Pipeline to Work on Next (13分50秒)
- 课时113：Summary and Thank You (4分41秒)
热门下载
热门帖子