Neural Networks: Where do I start?!?
summary
A 100‑level introduction explains neural nets as math that map inputs to outputs and focuses on five concrete parts: 5 components: inputs outputs weights biases hidden layers. Shows Keras (sitting on Keras on TensorFlow/Theano) examples: a tiny sequential model, compile/fit, use MSE loss and Adam optimizer. Demonstrations include a color triad, 72x72 grayscale cow images classifier, 6‑month windows to forecast month 7, anomaly detection for contaminated milk, regression for price per pound, and reinforcement learning for auction decisions.
AI-written summary · about
chapters (18)
- 0:00Start
- 0:54Why Am I Talking about Neural Networks
- 2:25Expectations
- 3:04What Is a Neural Network
- 7:29The Parts of a Neural Network
- 7:53Inputs and Outputs
- 11:59The Color Wheel
- 16:18Bias
- 16:55Hidden Layers
- 21:45Making a Neural Network
- 25:37Loss and Optimization
- 27:36Classifier
- 29:32Anomaly Detection
- 31:12Reinforcement Learning
- 32:00Reinforcement Learning
- 36:05Closing Thoughts
- 38:19Standardization
- 38:58Beginner Where Would I Recommend Getting Started on Reinforcement Learning
related talks
52mNeural Networks Walkthrough
58mNeural Networks: A Primer
01h 02mNeural Networks: A Primer
43mA Gentle Introduction to Neural Networks
01h 50mNeural Networks: Walkthrough
37mCode First, Math Later: Learning Neural Nets Through Implementation and Examples
24mSuper Accessible No Math Intro to Neural Networks For Beginners
50mMachine Learning Exposed: The Fundamentals