Unsupervised Learning

Question: What is unsupervised learning? This means the data is not labeled ahead of time for the algorithm. Used to analyze data to make sense…

Supervised Learning

Question: What is supervised learning? This means the data is pre-labeled for the algorithm. Helps when predicting from data we already identified. Example: Supervised machine…

Various Classification Algorithms

Question: What are the various classification algorithms? There are quite a few, so list the ones you know. There are subcategories you could mention, too.…

Deep Learning

Question: What do you mean by Deep Learning and why is it popular? Deep learning is a type of machine learning. It works like a…

Activation Function

Question: What is the role of activation function? Introduces non-linearity into a neural network to help it learn complex functions. Use examples if it helps.…

Auto-encoder

Question: What is an auto-encoder? Simply transforms unlabeled input into output. Makes output as close to input as possible within a neural network. Example: The…

Artificial Neural Network

Question: What are artificial neural networks? Algorithms mimic biological neural networks. They can adapt to changing input. Helps you redesign output criteria. Example: An artificial…

Structure of Neural Networks

Question: Describe the structure of artificial neural networks? It usually has a lot of inputs, often unlabeled. These are processed with a weighted sum and…

Gradient Descent

Question: Explain gradient descent. Gradient is how much an output changes by input. Measures the change in all weights with regard to change in error.…

Back Propagation

Question: What is the back propagation and how does it work? Training algorithm used for multilayer neural network. Moves the error from an end to…