Scribd
Upload
1 views2 pages

AI 2 Mark Answers

The document provides an overview of key concepts in Artificial Intelligence, including definitions of AI, adversarial search, uncertainty, and Bayes Rule. It outlines differences between supervised and unsupervised learning, describes ensemble learning and its types, and discusses applications of AI along with performance measurement. Additionally, it covers techniques to avoid overfitting, the logic behind Gaussian processes, and the use of Stochastic Gradient Descent in training neural networks.

Uploaded by

rsaksha2006
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
1 views2 pages

AI 2 Mark Answers

The document provides an overview of key concepts in Artificial Intelligence, including definitions of AI, adversarial search, uncertainty, and Bayes Rule. It outlines differences between supervised and unsupervised learning, describes ensemble learning and its types, and discusses applications of AI along with performance measurement. Additionally, it covers techniques to avoid overfitting, the logic behind Gaussian processes, and the use of Stochastic Gradient Descent in training neural networks.

Uploaded by

rsaksha2006
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 2

1.

Define Artificial Intelligence


AI is the field of computer science that enables machines to mimic human
intelligence such as learning, reasoning, and problem-solving.

2. What is Adversarial Search?


It is a search technique used in competitive environments (e.g., games) where
agents must consider the opponent’s moves to make optimal decisions (e.g., Minimax
algorithm).

3. Define Uncertainty
Uncertainty refers to incomplete or imperfect information about the environment or
future events, which affects decision-making in AI systems.

4. State Bayes Rule


P(A|B) = [P(B|A) * P(A)] / P(B)

5. Outline the differences between supervised learning and unsupervised learning


- Supervised Learning: Uses labeled data; learns from input-output pairs.
- Unsupervised Learning: Uses unlabeled data; discovers hidden patterns or
groupings.

6. What is a Random Forest?


A random forest is an ensemble method that builds multiple decision trees and
combines their outputs to improve classification or regression performance.

7. Define Ensemble Learning


Ensemble learning combines predictions from multiple models to achieve better
accuracy and robustness than any single model.

8. What is the significance of the Gaussian Mixture Model?


GMM models data as a mixture of several Gaussian distributions, useful in
probabilistic clustering and density estimation.

9. Draw the architecture of Multilayer Perceptrons


Multilayer Perceptron (MLP) consists of:
[Input Layer] → [Hidden Layers with Activation Functions] → [Output Layer]

10. Name any two activation functions


- ReLU: f(x) = max(0, x)
- Sigmoid: f(x) = 1 / (1 + e^-x)

11. What are the various applications of AI?


Applications include self-driving cars, virtual assistants, medical diagnosis,
recommendation systems, fraud detection, and robotics.

12. How will you measure the performance of AI applications?


Using evaluation metrics such as accuracy, precision, recall, F1-score, ROC-AUC,
confusion matrix, and MSE (for regression tasks).

13. Mention the needs of probabilistic reasoning in AI


Probabilistic reasoning is needed to handle uncertainty, incomplete data, and
probabilistic dependencies in decision-making processes.

14. Given that P(A) = 0.3, P(A|B) = 0.4, and P(B) = 0.5. Compute P(B|A)
P(B|A) = [P(A|B) * P(B)] / P(A) = (0.4 * 0.5) / 0.3 = 0.6667

15. How can overfitting be avoided?


- Use cross-validation
- Apply regularization (L1, L2)
- Prune decision trees
- Use dropout layers in neural networks
- Gather more training data

16. State the logic behind Gaussian Processes


Gaussian processes provide a probabilistic framework for learning distributions
over functions, useful for making predictions with uncertainty.

17. Write the three types of Ensemble Learning


- Bagging (e.g., Random Forest)
- Boosting (e.g., AdaBoost, XGBoost)
- Stacking (combining multiple models using a meta-model)

18. How is Expectation Maximization used in Gaussian Mixture Models?


EM iteratively performs:
- E-step: Estimate membership probabilities for each data point.
- M-step: Update the model parameters to maximize the likelihood.

19. What is Stochastic Gradient Descent and why is it used in training neural
networks?
SGD updates model weights using small batches or single samples, which makes
training faster and more memory-efficient for large datasets.

20. Why is ReLU better than Softmax? Give the equation for both.
ReLU helps avoid vanishing gradients and speeds up training in hidden layers, while
Softmax is used only in the output layer for classification.
- ReLU: f(x) = max(0, x)
- Softmax: f(x_i) = e^(x_i) / Σ e^(x_j)

You might also like