Artificial Intelligence · Unit 8
Machine Learning and Neural Networks
Exam-focused notes for Machine Learning and Neural Networks (Artificial Intelligence, BIT252): what the TU syllabus asks and how it has actually been tested, with 7 solved past questions from this unit.
What this unit covers
- Machine learning definition
- Artificial neural network mathematical model
- Neuron structure and activation functions
- Feed-forward neural networks
- Recurrent neural networks
- Back-propagation algorithm
- Learning rules and learning rates
- Neural network types and architectures
Back-propagation algorithm
What is learning rule?How learning is done in ANN using back propagation algorithm?[2+8]
--- A learning rule is a method or procedure that modifies the weights and biases of a neural network in order to improve its performance on a given task. It defines how the network learns from training data by updating connection weights based on the error...
Full solved answer →Why do we need recurrent neural network? How does back propagation learn to minimize the error? Explain[10]
Standard feedforward neural networks (FNN) have a fundamental limitation: they assume all inputs are independent of each other. Each input is processed in isolation, and the network has no memory of previous inputs. This creates problems for sequential data...
Full solved answer →Artificial neural network mathematical model
Describe the mathematical model of ANN. Differentiate feed-forward ANN from feed-back ANN. [2+3]
--- An Artificial Neural Network (ANN) is inspired by biological neurons. The mathematical model of a single artificial neuron consists of the following components: 1. Inputs and Weights: Each neuron receives n inputs x₁, x₂, ..., xₙ with corresponding weig...
Full solved answer →Express the mathematical model of a neuron. Distinguish between learning rule and learning rate. How Back-propagation algorithm is used in learning? Explain.[10]
Note: No specific curriculum notes were provided for this question. The answer below is based on standard, correct Artificial Neural Networks (ANN) theory as taught in BSc CSIT Neural Networks / AI courses. --- A biological neuron is mathematically modelled...
Full solved answer →Neural network types and architectures
Discuss different types of Neural Network. [5]
Note: Reference notes were not available for this topic. The following answer is based on standard Computer Science / AI curriculum content appropriate for BSc CSIT. --- A Neural Network is a computational model inspired by the human brain, consisting of in...
Full solved answer →Machine learning definition
What is machine learning? Describe about NLU and NLG. [5]
--- Machine Learning (ML) is a subfield of Artificial Intelligence (AI) that enables computers to learn from data and experience without being explicitly programmed for every task. Instead of writing fixed rules, a machine learning system: - Takes input dat...
Full solved answer →Neuron structure and activation functions
What is the task of activation function? What are its types? [5]
An activation function is a mathematical function applied to the output of each neuron in a neural network. Its main tasks are: 1. Introduces Non-linearity: Without an activation function, a neural network (no matter how many layers) would behave like a sim...
Full solved answer →Make Unit 8 stick
Practice BIT252 with flashcards & quizzes