Chapter 1 - Intro of ML Flashcards
What is the core definition of Machine Learning (ML)?
ML is a branch of Artificial Intelligence (AI) focused on building systems that can automatically learn and improve from experience without being explicitly programmed
According to Tom M. Mitchell, what constitutes machine learning?
A computer program learns from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E
What are the key components of machine learning?
data, algorithms, experience (training), and learning
List the general steps of how machine learning works
Data Collection, Data Preprocessing, Model Selection, Training, Testing, and Deployment
Give three real-world applications of machine learning.
Machine learning is applied in healthcare, finance, retail and technology. Some examples are predicting patient outcomes, fraud detection, personalized recommendations, and speech recognition
What are some of the benefits of using machine learning?
- automating complex tasks
- scalability to handle large datasets
- aptability as it improves with more data
- better decision making
What is supervised learning?
Supervised learning involves training a model on a labeled dataset, where the input data is paired with the correct output
Give an example of supervised learning
- Image classification
- spam detection
- predictive maintenance
- medical diagnosis
What is unsupervised learning?
Unsupervised learning works with unlabeled data. The model identifies patterns, groups, or associations within the dataset without predefined labels
What are some ways unsupervised learning is used?
Unsupervised learning is used for clustering, dimensionality reduction, customer segmentation, anomaly detection, and recommender systems
Explain reinforcement learning?
Reinforcement learning involves an agent learning to make decisions by interacting with an environment, receiving rewards or penalties. The goal is to maximize cumulative rewards over time
Provide an example of reinforcement learning in practice.
- game AI
- autonomous vehicles
- robotics
- personalized marketing
What is semi-supervised learning?
Semi-supervised learning is a combination of supervised and unsupervised learning, using a small amount of labeled data and a large amount of unlabeled data
When would you use semi-supervised learning?
Semi-supervised learning is useful in applications like speech recognition, web page classification and object detection when labeled data is scarce or expensive to obtain
How do biological neural networks work?
They consist of interconnected neurons that transmit electrical and chemical signals to process information
What are the key components of a biological neuron?
The key parts are the dendrites, cell body (soma), axon, and synapse
What are Artificial Neural Networks (ANNs)?
ANNs are computational models inspired by biological neurons, composed of artificial neurons arranged in layers that learn to recognize patterns in data
What are the key parts of an artificial neuron (perceptron)?
The key parts are inputs, weights, summation function, activation function, and output
What part of an artificial neuron is analogous to the dendrites of a biological neuron?
The inputs of an artificial neuron are analogous to the dendrites of a biological neuron
How do the functions of a biological neuron’s cell body and artificial neuron’s summation function relate?
Both perform the initial processing of input signals
What is the function of weights in an artificial neural network?
Weights determine the importance of each input in the network. They influence the strength and importance of each input
What is the function of an activation function in an artificial neuron?
The activation function decides whether to activate the neuron based on the combined input. It determines whether a neuron will “fire” based on stimuli
How does learning happen in biological neural networks compared to artificial neural networks?
Biological networks learn through strengthening or weakening synapses (synaptic plasticity), while artificial networks learn by adjusting the weights during training using algorithms like backpropagation
What are some key differences between biological and artificial neural networks?
Biological networks communicate using electrical and chemical signals and are highly complex and adaptable. Artificial networks communicate through mathematical computations, are task-specific, and require large amounts of data
In an image recognition task, what do the inputs to an artificial neuron represent?
In an image recognition task, the inputs could be pixel values of an image
