Advanced AI and Machine Learning Fundamentals

Computer Science

A rigorous assessment of technical concepts in artificial intelligence, covering neural networks, data processing, and algorithmic theory.

AI Machine Learning Data Science Algorithms

15 Questions Hard Ages 15+ Sep 20, 2026

Practice Test PDF Answer Key PDF

Choose a Study Format

Quiz

Test your knowledge with multiple-choice questions and instant feedback.

Flashcards

Flip through cards to review terms, definitions and key concepts.

True or False

Quickly decide if statements are true or false to reinforce understanding.

Match Game

Race against time to match questions with their correct answers.

Study Guide

Read through all topics and answers as a comprehensive study reference.

Listen Mode

Listen to questions and answers read aloud — great for audio learners on the go.

Embed This Study Set

Add this interactive study set to your website or blog — all 6 formats included.

<div data-quixly-id="7451"></div>
<script src="https://www.quixlylearn.com/assets/embed/widget.js"></script>

Learn more about embedding →

About this Study Set

This study set covers Computer Science through 15 practice questions. A rigorous assessment of technical concepts in artificial intelligence, covering neural networks, data processing, and algorithmic theory. Every question includes the correct answer so you can learn as you go — pick any format above to get started.

Questions & Answers

Browse all 15 questions from the Advanced AI and Machine Learning Fundamentals study set below. Each question shows the correct answer — select a study format above to practice interactively.

1 Which activation function is defined by the mathematical expression f(x) = max(0, x) and is widely used to mitigate the vanishing gradient problem in deep neural networks?

A Sigmoid
B Tanh
C Rectified Linear Unit (ReLU)
D Softmax

2 In the context of supervised learning, what is the specific term for the phenomenon where a model performs exceptionally well on training data but fails to generalize to unseen data?

A Underfitting
B Overfitting
C Bias-Variance Tradeoff
D Gradient Descent

3 Which mathematical theorem provides the basis for updating the probability of a hypothesis as more evidence or information becomes available?

A Bayes' Theorem
B Central Limit Theorem
C Pythagorean Theorem
D Fundamental Theorem of Calculus

4 In a Convolutional Neural Network (CNN), what is the primary purpose of a 'pooling layer'?

A To increase the number of trainable parameters
B To perform non-linear activation
C To reduce the spatial dimensions of the input volume
D To normalize the weights of the network

5 What does the 'k' represent in the K-Nearest Neighbors (KNN) algorithm?

A The number of iterations for training
B The number of nearest data points to consider for classification
C The constant factor in the learning rate
D The total number of clusters in the dataset

6 Which optimization algorithm is an extension of Stochastic Gradient Descent that utilizes the moving average of gradients to accelerate convergence?

A Backpropagation
B Adam (Adaptive Moment Estimation)
C Principal Component Analysis
D Linear Regression

7 What is the specific role of 'Dropout' in a deep neural network during the training process?

A To increase the learning rate
B To prevent overfitting by randomly disabling neurons
C To convert input data into a vector format
D To reduce the time required for data pre-processing

8 Which term describes the process of transforming raw data into a format suitable for machine learning algorithms, often involving scaling and encoding?

A Feature Engineering
B Backpropagation
C Supervised Learning
D Reinforcement Learning

9 In reinforcement learning, what does the 'exploration vs. exploitation' trade-off refer to?

A Choosing between supervised and unsupervised learning
B Deciding between trying new actions or choosing known high-reward actions
C The balance between CPU and GPU utilization
D The trade-off between training speed and model accuracy

10 What is the primary function of the 'Softmax' function when used in the output layer of a multi-class classification neural network?

A To normalize the output into a probability distribution
B To transform negative values into zero
C To increase the weight of hidden neurons
D To calculate the loss function

11 Which type of machine learning model is specifically designed to handle sequential data, such as time series or natural language, by maintaining a hidden state?

A Recurrent Neural Network (RNN)
B Decision Tree
C Random Forest
D Support Vector Machine

12 In data science, what is the 'Curse of Dimensionality'?

A The inability of computers to process large datasets
B The observation that data becomes sparse as the number of features increases
C The requirement for more memory to store high-dimensional arrays
D The difficulty of visualizing more than three dimensions

13 What is the purpose of the 'learning rate' hyperparameter in gradient-based optimization?

A To define the depth of the neural network
B To determine the step size at each iteration while moving toward a minimum
C To select the number of epochs for training
D To initialize the weights of the model

14 Which clustering algorithm relies on the calculation of centroids and iteratively assigns data points to the nearest cluster center?

A K-Means
B Logistic Regression
C Naive Bayes
D Linear Discriminant Analysis

15 In unsupervised learning, which technique is used to reduce the number of variables in a dataset while retaining as much variance as possible?

A Principal Component Analysis (PCA)
B Gradient Boosting
C Cross-Validation
D Data Augmentation

📱

Study on the go

Download Quixly and access all study formats on your phone — anywhere, anytime.