Artificial Intelligence Flashcards
What are some key areas where AI is applied?
Knowledge representation
Planning
Natural language processing
Machine learning
Computer vision
Robotics
What are the two main types of AI?
General-purpose AI: Like the robots in science fiction, capable of performing any task.
Special-purpose AI: Designed for specific tasks like playing chess, speech recognition, and fraud detection.
What are the key differences between AI and humans?
AI follows pre-programmed goals, whereas humans set their own goals.
AI can share knowledge instantly, while humans learn individually.
AI’s actions can be restricted, while humans can make any choice.
What is Machine Learning (ML)?
A subset of AI that enables systems to learn from data and improve over time without being explicitly programmed.
What is Deep Learning (DL)?
A specialized branch of ML that uses neural networks with multiple layers to learn from vast amounts of data.
What is the difference between Machine Learning and Deep Learning?
Machine Learning: Works with structured data and requires feature engineering.
Deep Learning: Works with unstructured data and learns features automatically.
What are Neural Networks?
Computational models inspired by biological neural networks that process data through layers of interconnected nodes (neurons).
What are the three layers of a Neural Network?
- Input Layer – Receives raw data
- Hidden Layers – Processes information using weights and activation functions
- Output Layer – Produces the final prediction or classification
What are common Activation Functions in Neural Networks?
Sigmoid: Outputs values between 0 and 1
Tanh: Outputs values between -1 and 1
ReLU (Rectified Linear Unit): Outputs max(0, x), helping to prevent vanishing gradients
Softmax: Converts outputs into probability distributions
What is Backpropagation?
A method used in training neural networks by adjusting weights based on the error between predicted and actual outputs.
What are common Gradient Descent methods for optimizing neural networks?
Stochastic Gradient Descent (SGD)
Mini-batch Gradient Descent
Batch Gradient Descent
What are Vanishing and Exploding Gradient Problems?
Vanishing Gradient: Small gradients cause learning to slow down or stop in deep networks.
Exploding Gradient: Large gradients cause instability and divergence in training.
What are Convolutional Neural Networks (CNNs)?
A type of deep neural network designed for image processing, consisting of convolutional layers that detect spatial features.
What are the key layers in a CNN?
- Convolutional Layer – Extracts features from input images
- ReLU Activation – Adds non-linearity
- Pooling Layer – Reduces dimensionality and computation
- Fully Connected Layer – Outputs final classification
What is the role of Filters (Kernels) in CNNs?
Filters scan through the image, detecting patterns such as edges, textures, and shapes.
What is Pooling in CNNs?
A technique that reduces the spatial size of feature maps while retaining important information (e.g., max pooling, average pooling).
