AI Basics
What is Artificial Intelligence?
What is Artificial Intelligence?
Define supervised and unsupervised learning.
Supervised: Labeled data; Unsupervised: No labels, patterns discovered.
Who invented the perceptron model?
Frank Rosenblatt.
Name one use of AI in the healthcare sector.
Disease prediction, diagnostics, or robot-assisted surgery.
What is algorithmic bias in AI?
When an AI system reflects human or systemic biases due to biased training data.
Name the three main types of AI (based on capability).
Narrow AI, General AI, and Super AI.
What is reinforcement learning? Give an example.
Learning by trial and error via rewards. Example: A robot learning to walk.
Name three basic components of a perceptron.
Input nodes, weights and bias, activation function.
How does AI help in self-driving cars?
Processes sensor data to detect lanes, obstacles, and make driving decisions.
Why is explainability important in AI models?
To help users understand decisions, ensure fairness, and build trust.
What is the Turing Test used for?
To evaluate whether a machine's behavior is indistinguishable from a human's.
What is overfitting in machine learning?
A model learns training data too well, including noise, and performs poorly on new data.
What is the perceptron activation function formula?
f(x) = 1 if w·x + b > 0, else 0.
How is AI used in chatbots and virtual assistants?
For understanding natural language, intent recognition, and generating responses.
What are the ethical concerns of AI in surveillance?
Invasion of privacy, misuse for social control, lack of consent.
Explain the difference between Strong AI and Weak AI.
Strong AI has human-level consciousness; Weak AI is task-specific and lacks consciousness.
Differentiate between classification and regression.
Classification: Predicting categories; Regression: Predicting continuous values.
Differentiate between single-layer and multi-layer perceptron.
Single-layer: one layer, handles linear problems. Multi-layer: multiple layers, handles non-linear problems.
Name 2 AI-based applications used in the education industry.
AI tutors, automated grading, learning analytics
How does AI affect employment and jobs?
Automates repetitive jobs, creates new tech roles, but may cause job displacement.
Describe the goals and subfields of AI.
Goals: Learning, reasoning, self-correction. Subfields: ML, NLP, robotics, vision, planning.
Explain how K-Means Clustering works.
Divides data into K clusters by minimizing intra-cluster variance using centroids.
Explain forward and backward propagation in a neural network.
Forward: input → output; Backward: error sent back to adjust weights (via backpropagation).
Describe how NLP is applied in sentiment analysis.
Analyzes text to determine emotion (positive/negative/neutral) using classification algorithms.
Define the term "AI transparency" and give an example.
Making AI decision-making understandable. Example: Explaining why a loan was denied.