20250331_215934.heic

# Machine Learning ## Chapter 1: What is Machine Learning? ### What is Machine Learning? * Machine learning is about **making computers learn from data**. * Arthur Samuel (1959): Machine Learning: field of study that gives computers the ability to learn without being explicitly programmed. *...

# Machine Learning ## Chapter 1: What is Machine Learning? ### What is Machine Learning? * Machine learning is about **making computers learn from data**. * Arthur Samuel (1959): Machine Learning: field of study that gives computers the ability to learn without being explicitly programmed. * Tom Mitchell (1997): Well-posed Learning Problem: A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance at T, as measured by P, improves with experience E. ### Example: * Task: T = playing checkers. * Performance Measure: P = percent of games won against opponents. * Experience: E = playing practice games against itself. ### Types of Machine Learning Systems * Supervised learning * Unsupervised learning * Semisupervised learning * Reinforcement Learning #### Supervised learning * **Learn from labeled Data** * In supervised learning, the training data you feed to the algorithm includes the desired solutions, called **labels**. * A **classification** problem (spam filter) * A **regression** problem (predicting house prices) #### Unsupervised learning * **Learn from unlabeled Data** * The training data is unlabeled. * **Clustering** * **Dimensionality reduction** * **Anomaly detection** #### Semisupervised learning * **Learn from partially labeled Data** * Deap Belief Networks (DBNs) #### Reinforcement Learning * The learning system, called an **agent**, can observe the environment, select and perform actions, and get **rewards** in return. * **Policy**: defines what action the agent should choose when it is in a given situation. * The agent will learn by itself what is the best policy to get the most reward over time. ### Main Challenges of Machine Learning * **Insufficient Quantity of Training Data** * **Nonrepresentative Training Data** * Sampling bias * **Poor Quality Data** * Outliers * Missing values * **Irrelevant Features** * Feature selection * Feature extraction * **Overfitting the Training Data** * Regularization * **Underfitting the Training Data** * Selecting a more powerful model * Feeding better features to the learning algorithm * Reducing the constraints on the model