About the Course
Perceptrons, Neurons, Backpropagation, Multi-layer, Hyperparamaters, CNN and RNN
- Neural Network Architecture
- In this chapter you will learn about the architectures of AI and where does Machine Learning and Deep Learning lie.
- Single Layer Perceptron
- In this chapter the basic neuron architecture is discussed.
- Aggregation Function
- Input to the neuron is achieved by aggregating data or input from the external world. The purpose of the aggregation function is combine the inputs and the weights.
- Activation Function
- The aggregated input is fed into an activation function which give the output of the neuron. The type of problem decides the nature of the activation function.
- Implementing Boolean Logic
- The neuron has been used to implement basic Boolean logic such as OR, NOT, NAND and AND gates. In this chapter we see how to implement this.
- Backward Propagation
- In a multi-layer perceptron architecture the weights are updated using back propagation. In this chapter this method is dicussed.