Chapter 14 Flashcards
Artificial Intelligence (AI)
Focused on studying human thought processes and recreating them with machines like computers
Strong AI
Hypothetical AI that matches or exceeds human intelligence and can perform any intellectual task that humans can
Weak AI (Narrow AI)
AI that performs a specific function, previously requiring human intelligence, and does so at human levels or better
Algorithm
A set of clear steps to solve a problem or complete a task.
Machine Learning (ML)
A type of artificial intelligence that allows systems to automatically learn and improve from experience without being explicitly programmed
Supervised Learning
Giving the system data and expected outcome results
Classification
When a computer learns to put things into groups based on their information
Supervised Learning Classifications
Binary Classification
Multi-Class Classification
Multi-Label Classification
Imbalanced Classification
Binary Classification
Only two possible groups for the data
Multi-Class Classification
More than two groups for the data
Multi-Label Classification
Each example can belong to more than one group at the same time
Imbalanced Classification
Some groups (classes) have way more examples than others
Semi-Supervised Learning
Giving small amounts of labelled data and large amount of unlabelled data
Unsupervised Learning
Computer finds patterns in data on its own without labels or much help
Reinforcement Learning
System learns to achieve a goal in an uncertain and complex environment
Deep Learning
Artificial neural networks learn from large amounts of data
Neural Network (NN)
Virtual “neurons” arranged in layers that work together to simulate the human brain, to solve problems or recognize patterns
Node
Part of Neural Network that takes input, processes it, and produces an output
Backpropagation
A supervised learning method used to update a model’s parameters to improve the accuracy of its predictions.
Recurrent Neural Network (RNN)
A type of neural network that processes sequential or time-series data, where the network’s decision depends on previous outputs
Convolutional Neural Network (CNN)
Neural network used to understand images by looking at parts like edges, curves, and colors, and then putting them together to figure out what the image shows
Generative Adversarial Network (GAN)
Has 2 parts
Generator: Learns to create realistic data.
Discriminator: Learns to distinguish between real and generated (fake) data
Computer Vision
The ability of systems to recognize objects, scenes, and activities in images
Natural Language Processing (NLP)
The ability of systems to understand and process text the way humans do
Speech Recognition
The ability of systems to automatically and accurately transcribe human speech
Chatbot
A program that uses AI and natural language processing to simulate human conversation, either by voice or text
Machine Learning Systems
Systems that perform new tasks accurately by learning from training data or historical data with known labels
