Challenges of developing artificial intelligence Reasoning, problem-solving Knowledge Representation Machine learning Natural language processing Perception Social intelligence General intelligence 1. Machine Learning Types of Machine Learning 1. Supervised Learning 2. Unsupervised Learning 3. Semi-supervised Learning (SSL) 4. Reinforcement Learning Economic Benefits of Machine Learning 2. Deep learning 3. Neural Networks Feedforward Neural Networks 4. Cognitive Computing 5. Natural Language Processing 6. Computer Vision Other Branches of Artificial Intelligence 1. Robotics 2. Expert Systems 3. Fuzzy Logic “Strong” versus “Weak” AI Weak artificial intelligence Strong artificial intelligence Differences between strong and weak AI in a nutshell The Chinese Room Argument Against “Strong AI” The Gödelian Argument Against “Strong AI” Search and problem solving Search in practice: getting from A to B Toy problem: chicken crossing The easy version of the rowboat puzzle States of the chicken crossing puzzle State space, transitions, and costs The state space Transitions Costs Problem Solving Techniques Heuristics Example: Travelling Salesman Problem Searching Algorithms Types of Searching Algorithms Informed Search Types of informed Search Algorithms Uninformed Search Types of Uninformed Searching Algorithms Evolutionary Computation Genetic Algorithms Problem Formulation in AI Problem-solving agents in artificial intelligence Game Playing in Artificial Intelligence Minimax Algorithm in Game Theory Likelihood ratio Application Example 2: Classification Don't get carried away with your machine learning algorithm The nearest neighbour classifier Regression Coefficients or weights Applications Logistic Regression But Machine Learning is not the answer to everything Deep learning How to build neural networks Weights and inputs Activations and outputs How neurons activate Perceptron Networks Neural Network Classifier: An Example Convolutional neural networks (CNNs) Generative adversarial networks (GANs)