Advanced Machine Learning Flashcards
What is Deep Learning?
Deep learning is a subset of machine learning that has become increasingly popular in recent years due to its performance along with improved computer speed.
What is the process of neural networks?
- Take our inputs and multiply each input by a weight.
- Add the results from step 1 together.
- Add a value (called bias) to the result from step 2
- Pass result from step 3 through an activation function.
- The result from step 4 is then passed to the next layer, where the processed is repeated.
What is Node?
One unit that process incoming information and produces a single output number
What is layer?
A collection of nodes working in parallel
What is Input layer?
The first set of nodes that work on the features of the input sample
What is Output layer?
The last set of nodes that output a prediction
What is Forward Propagation?
The processing and passing of information forward through all layers to produce an output prediction
What is cost function?
The difference between the predictions of all samples and their true labels
What is Backward Propagation?
The process of updating the weights of each node to reduce the cost function
What is Epoch?
The process of completing one forward Propagation step on each sample in the training set and updating the weights of each node with backward Propagation is called an epoch
What is activation functions?
They can create non-linear functions to predict class or values
What is sigmoid?
The sigmoid function maps all output values of the node to a value between 0 and 1
What is Tanh (or hyperbolic tangent)?
Tanh maps the output of a node to a value between -1 and 1, as shown above on the y axis of the second plot
Wha is ReLu (Rectified Linear Unit)?
It maps all negative outputs from a node to 0 and returns all positive outputs as is (linear function)
What is Output layer Activation functions?
The exception is the output layer. Specific activation functions are required for a model to produce specific kinds of outputs
What is Linear (regression models)?
If the model should have a continuous output that can produce positive or negative. A Linear activation function simply returns the value that was passed to it. g(z) =z.
What is sigmoid (binary classification)?
If the model should only output 0 or a 1, such as in binary classification problems
What is softtmax (multiclass classification models)?
If the model should return the most likely candidate from a finite list of options, such as a multiclass classification problem
What is ythat?
The prediction of a model for all samples in X.
What is cost?
The combination of all of a model’s errors on all samples
What is cost function or loss function?
The difference between the predictions of all samples and their true labels
What is Gradient Descent?
Changing weights to reduce the cost function
What is Keras?
Keras is an API that acts as an interface for tensorflow, a popular choice for building networks, especially as you are first learning
What is Bias?
Is is related to underfitting
What is Variance?
Is it related to overfitting
What is CNN?
Convolutional Nerual Networks (CNNs) revolutionized computer vision because of how well they do at working with image data.
