Machine Learning Exposed Introduction to Machine Learning
summary
Covers supervised, unsupervised and reinforcement methods with concrete demos: linear regression on 10 examples of house size yields a squared error cost 4418 and uses gradient descent with learning rate alpha; k‑means clustering with centroids and Euclidean distance; iris classification on iris 150 samples and MNIST via convolutional nets; shows a Java/Spring + Angular2 WebSocket visualization (GitHub) wired to DeepLearningForge that trains weights and predicts Setosa at ~0.999 confidence; mentions TensorFlow/TensorBoard, GANs that generate realistic faces and a Tesla self‑driving sensor demo.
AI-written summary · about
chapters (13)
- 0:00Start
- 4:48Self-driving cars
- 8:26Reinforcement Learning
- 10:27The hypothesis
- 12:48What's the error?
- 12:58Squared error (or the Cost Function)
- 15:26How theta changes the Cost Function
- 20:27Polynomial Regression
- 20:44Gradient Descent with multiple variables
- 21:47Limits of linear regression
- 29:25Unsupervised Learning Let's dive in now!
- 37:02K means algorithm
- 37:30What's the cost?
related talks
49mMachine Learning Exposed: Introduction to Machine Learning
50mMachine Learning Exposed: The Fundamentals
44mMachine Learning Exposed
55mTensorflow: Learning Machines
54mMachine Learning explained with Examples
54mMachine Learning for Developers
26mHow Machine Learning Can Help
50mBuilding Intelligent Mobile Apps using Machine Learning