ARTIFICIAL INTELLIGENCE (NCS-702)
B.Tech – Semester VII
SECTION A
(Attempt all questions)
(a) Learning agent with architecture
A learning agent is an intelligent agent that has the ability to improve its performance over time by learning from its experiences and feedback from the environment. The architecture of a learning agent consists of four main components: the performance element, learning element, critic, and problem generator. The performance element selects actions based on current knowledge, while the learning element modifies the performance element to improve future behavior. The critic evaluates how well the agent is performing by comparing outcomes with desired goals, and the problem generator suggests new actions that help the agent gain useful experience. This architecture enables continuous learning and adaptation.
(b) Computer Vision
Computer Vision is a branch of artificial intelligence that enables machines to interpret, analyze, and understand visual information from the real world such as images and videos. It involves techniques for image acquisition, processing, feature extraction, and object recognition. Computer vision systems are widely used in applications such as facial recognition, medical imaging, autonomous vehicles, surveillance systems, and industrial inspection.
(c) Time and space complexity of DFS search strategy
Depth First Search is a search strategy that explores the deepest nodes of a search tree first before backtracking. The time complexity of DFS depends on the branching factor and the maximum depth of the tree and is generally expressed as O(bm)O(b^m)O(bm), where bbb is the branching factor and mmm is the maximum depth. The space complexity of DFS is relatively low compared to BFS, as it stores only a single path from the root to a leaf node along with sibling nodes, making it O(bm)O(bm)O(bm).
(d) Soundness property of inference
Soundness is a property of an inference system which guarantees that any sentence derived by the inference rules is logically entailed by the knowledge base. In other words, a sound inference system never derives false conclusions. Soundness is essential in artificial intelligence systems to ensure correctness and reliability of reasoning and decision-making.
(e) PEAS description for a Satellite Agent
PEAS stands for Performance measure, Environment, Actuators, and Sensors. For a satellite agent, the performance measure includes successful data transmission, fuel efficiency, and mission completion. The environment consists of space conditions such as vacuum, radiation, and orbital paths. Actuators include thrusters, solar panel controllers, and communication antennas. Sensors include cameras, gyroscopes, temperature sensors, and signal receivers. This PEAS description helps define the task environment clearly.
(f) Application areas of Machine Learning
Machine learning is widely applied in areas such as speech recognition, image processing, recommendation systems, medical diagnosis, financial forecasting, fraud detection, and autonomous systems. It enables systems to learn patterns from data and make intelligent predictions or decisions without explicit programming.
(g) Supervised and Unsupervised Learning
Supervised learning is a machine learning approach where the model is trained using labeled data, meaning the input and corresponding output are known. It is commonly used for classification and regression problems. Unsupervised learning, on the other hand, deals with unlabeled data and aims to discover hidden patterns or structures within the data. Clustering and dimensionality reduction are typical examples of unsupervised learning.
SECTION B / SECTION C (Long Answer)
(Attempt any one)
(a) Dimensionality reduction and Principal Component Analysis (PCA)
Dimensionality reduction is the process of reducing the number of input variables in a dataset while preserving as much important information as possible. High-dimensional data increases computational complexity and may degrade model performance due to the curse of dimensionality. Principal Component Analysis (PCA) is a widely used dimensionality reduction technique that transforms the original correlated variables into a new set of uncorrelated variables called principal components. These components are ordered such that the first few retain most of the variance present in the original data. PCA improves computational efficiency, reduces noise, and enhances data visualization.
(b) Bayesian Theory and Bayesian Classification
Bayesian theory is based on Bayes’ theorem, which provides a mathematical framework for updating probabilities based on new evidence. In Bayesian classification, prior probability represents the initial belief about a class before observing data, while posterior probability represents the updated belief after considering the evidence. Likelihood measures how probable the observed data is given a class. Bayesian classifiers use these probabilities to assign the most probable class to a given instance. This approach is particularly useful in situations involving uncertainty and incomplete information.
