Deep Learning
Deep Learning for Computer Vision
Learning types:
- Reinforcement learning - learn to select an action of maximum payoff
- Supervised learning - given input, predict output.
2
types: regression (continuous values), classification (labels) - Unsupervised learning - discover internal representation of an input (also includes self-supervised learning)
Artificial neuron representation:
- input vector (“synapses”) is just the given features - weight vector which regulates the importance of each input - bias which adjusts the weighted values, i.e., shifts them - net input vector which is linear combination of inputs - activation function through which the net input is passed to introduce non-linearity - the activation vector which is the neuron output vector
Artificial neural network representation:
- Each neuron receives inputs from inputs neurons and sends activations to output neurons
- There are multiple neuron layers and the more there are, the more powerful the network is (usually)
- The weights learn to adapt to the required task to produce the best results based on some loss function
Popular activation functions - ReLU, sigmoid and softmax, the last two of which are mainly used in the last layer before the error function:
Popular error functions - MSE (for regression), Cross Entropy (for classification):
Backpropagation - weight update algorithm during which the gradient of the error function with respect to the parameters