Supervised Learning - Classification Week 9 Challenge
-
A neural network is a type of machine learning algorithm that is designed to simulate the behavior of the human brain. It is made up of a large number of interconnected processing nodes, called neurons, which work together to process and analyze data. Neural networks are typically used for tasks such as pattern recognition, classification, and prediction.
-
Deep learning is a subfield of machine learning that involves training neural networks with multiple layers to learn and extract features from complex data. These multiple layers of neural networks are what make it "deep". Deep learning algorithms have achieved state-of-the-art performance in various tasks such as image and speech recognition, natural language processing, and game playing.
-
The activation function in a neural network is responsible for introducing non-linearity into the output of each neuron. Without an activation function, the output of a neuron would simply be a linear combination of its input. By applying an activation function, the output can be transformed into a non-linear function that can capture more complex patterns in the data.
-
There are several types of activation functions, including:
- Sigmoid function: maps any input value to a value between 0 and 1
- ReLU (Rectified Linear Unit) function: sets any negative input value to zero, and leaves any positive input value unchanged
- Tanh function: similar to the sigmoid function, but maps input values to a range between -1 and 1
- Softmax function: used in the output layer of a neural network for multiclass classification tasks, as it ensures that the outputs sum to 1 and represent the probabilities of each class.
- Backpropagation is an algorithm used to train neural networks by adjusting the weights of the connections between neurons. It involves computing the gradient of the loss function with respect to the weights of the network, and then using this gradient to update the weights in the opposite direction to the gradient. This process is repeated iteratively until the network reaches a state of minimum loss, at which point it has learned to accurately predict the output for a given input.