Neural Networks Revision Notes Flashcards
What is a Neural Network in machine learning?
A Neural Network is a series of algorithms that aim to recognize underlying relationships in a set of data through a process that mimics the way the human brain operates. It is a core component of deep learning
What is an Artificial Neuron?
An artificial neuron is a computational model inspired by the neurons in the human brain. It receives input, processes it with a set of weights and a bias, and uses an activation function to output a result.
What are weights and biases in a Neural Network?
Weights are the coefficients of the input variables; they represent the strength of the connection between units. Biases are additional constants added to the input to influence the output; they allow the model to fit better with the data.
What is an Activation Function?
An activation function in a neural network model is a mathematical gate in between the input feeding the current neuron and its output going to the next layer. It decides whether a neuron should be activated or not.
Implications neural networks
Technology: Powering voice and facial recognition systems, enhancing user interactions with AI assistants.
Healthcare: Assisting in diagnosis, personalizing treatment plans, and predicting disease outbreaks.
Finance: Detecting fraudulent transactions, automating trading, and managing credit risk.
Automotive: Advancing autonomous driving systems and improving safety features.
Manufacturing: Optimizing supply chains, predicting equipment failures, and automating quality control.
Entertainment: Generating realistic animations and effects, personalizing content recommendations.
Retail: Tailoring customer experiences, managing inventory, and optimizing logistics.
What are the different layers in a Neural Network?
Input Layer: The layer that receives the input signal to be processed.
Hidden Layer(s): Intermediate layer(s) that perform computations and feature extraction.
Output Layer: The final layer that produces the output of the model.
How do neural networks mimic the brain
Structure: They are composed of units called artificial neurons, arranged in layers, which resemble the interconnected neuron cells in the human brain.
Signal Processing: Just as neurons transmit signals through synapses in the brain, artificial neurons process and pass signals to other neurons within the network.
Learning: Neural networks learn from examples in a way that’s inspired by the human learning process. The network adjusts the weights (synaptic strengths) based on the input data it receives, similar to how synaptic strengths in the brain are believed to change during learning (synaptic plasticity).
Parallel Processing: They can process multiple pieces of information simultaneously, akin to parallel processing in the brain.
Adaptability: Neural networks can adapt their weights (connections) to new data without being explicitly programmed, paralleling the brain’s ability to adapt and learn from new experiences.
