Artificial Intelligence Flashcards
True or False: Machine Learning is a subset of Artificial Intelligence.
True
Machine learning (ML) = algorithms which allow computers to learn without explicit programming
Fill in the blank: The ability of a computer or machine to perform tasks commonly associated with intelligent beings is known as ____
Artificial Intelligence
May also be known as the ability to mimic human behavior?
What is the primary goal of Natural Language Processing (NLP)?
To enable machines to understand and interpret human language.
Which of the following is NOT a type of machine learning?
(A) Supervised Learning
(B) Unsupervised Learning
(C) Reinforcement Learning
(D) Structural Learning
D) Structural Learning
True or False: Deep Learning is a more complex form of Machine Learning that uses multiple layers of neural networks.
True
What is the Turing Test used for?
To assess a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human.
Fill in the blank: AI that can perform tasks without human intervention is referred to as ______.
Autonomous AI
Which of the following ML’s will yield results that are inconsistently accurate? One that has
A. Low bias, low variance
B. Low bias, high variance
C. High bias, low variance
D. High bias, high variance
And what does this mean about the level of fit?
Answer: B
Bias = measure of inaccuracy of the ML model’s performance after training
Variance = measure of imprecision (lack of reproducibility). The square of the standard deviation. Due to sensitivity to small fluctuations in the training set.
Keep in mind the Bias-Variance Trade-Off
Low bias, low variance = ideal; accurate and precise/reproducible; consistently accurate results
Low bias, high variance = inconsistently accurate results. This can mean that the model has been overfit to the data.
High bias, low variance = consistently inaccurate results. This often suggests that the model has been underfit to the data.
High bias, high variance = very weak model that does not perform well at all.
True/False: the goal of any machine learning model is to achieve optimum model complexity by reducing both bias and variance to their lowest points.
Answer: TRUE
This is a challenge due to the Bias-Variance Trade-Off.
Total error = (bias squared) + variance + irreducible error
Which of the following ML algorithms is best for unsupervised learning?
A. Classification methods
B. Regression methods
C. Clustering methods
D. Ensemble methods
C. Clustering methods are appropriate learning algorithms for unstructured machine learning.
Classification methods, regression methods, and ensemble methods are intended for structured ML.
Remember that unsupervised algorithms can be used with supervised learning. But not vice versa.
Two other algorithms for unsupervised learning include association rules and dimensionality reduction methods.
Neural networks can be used for supervised or unsupervised learning.
What is the primary goal of supervised learning?
To learn a mapping from inputs to outputs using labeled training data.
True or False: Unsupervised learning requires labeled data.
False
Fill in the blank: In reinforcement learning, an agent learns to make decisions by receiving ______ from the environment.
rewards
What is overfitting in the context of machine learning?
When a model learns the training data too well, including noise, leading to poor generalization on unseen data.
Which algorithm is commonly used for classification tasks?
Support Vector Machine (SVM)
