INSIGHT X

AI ESCAPE ROOM
No of Rounds : 2
LEVEL 1 - Multiple Choice Question (MCQ) Round [20 Questions]
Terms & Conditions:

• Qualification: Teams must score 50% to advance.

• Online Mode: Quiz conducted online.
• Time Limit: 45 seconds per question.
• Bonus Points: Sequence of 3 correct answers gets bonus.

1. Who is Google's father of AI?

A) Andrew Ng
B) Geoffrey Hinton
C) Yann LeCun
D) Arthur Samu
Answer: B) Geoffrey Hinton

2. Who is often referred to as the "Father of Machine Learning" for his pioneering work in the
field?

A) Andrew Ng
B) Geoffrey Hinton
C) Yann LeCun
D) Arthur Samuel
Answer: D) Arthur Samuel (Correct)

3. Which term refers to a type of neural network layer that connects every neuron to every neuron
in the subsequent layer?

A) Convolutional Layer
B) Recurrent Layer
C) Fully Connected Layer
D) Dropout Layer
Answer: C) Fully Connected Layer (Correct)
4. What is "Transfer Learning" in deep learning?

A) The process of training models from scratch

B) Using pre-trained models and fine-tuning them for specific tasks

C) Combining multiple models to improve accuracy

D) The process of converting deep models to shallow models

Answer: B) Using pre-trained models and fine-tuning them for specific tasks

5. Which concept involves training multiple AI models simultaneously to improve the overall
performance of each model?

A) Transfer Learning

B) Reinforcement Learning

C) Ensemble Learning

D) Active Learning

Answer: C) Ensemble Learning (Correct)

6. What is the concept of "Generative Adversarial Networks" (GANs) used for in AI?

A) Identifying fraudulent activities in financial transactions

B) Enhancing the interpretability of AI models

C) Generating realistic data samples and images

D) Solving optimization problems

Answer: C) Generating realistic data samples and images

7. Which AI-powered virtual assistant is known for its ability to understand and generate human-
like language?

A) Alexa (Amazon)

B) Siri (Apple)

C) Cortana (Microsoft)

D) Bixby (Samsung)

Answer: B) Siri (Apple)

8. What is the concept of "Edge AI" aimed at achieving in AI systems?

A) Moving AI computations to the cloud for better scalability

B) Increasing energy consumption of AI devices

C) Enabling AI computations to be performed on local devices

D) Focusing on AI algorithms for quantum computers

Answer: C) Enabling AI computations to be performed on local devices

9. Which AI concept involves imitating the human brain's interconnected neurons and synapses to
build intelligent systems?

A) Artificial Neural Networks (ANNs)

B) Genetic Algorithms

C) Support Vector Machines (SVMs)

D) Markov Chain Monte Carlo (MCMC)

Answer: A) Artificial Neural Networks (ANNs) (Correct)

10. What does the "R" stand for in the acronym "RNN" in the context of neural networks?

A) Recursive

B) Recursive

C) Recurrent

D) Robotic

Answer: C) Recurrent (Correct)

11. What is the term for a type of neural network architecture inspired by the organization of the
human brain?

A) Convolutional Neural Network (CNN)

B) Recurrent Neural Network (RNN)

C) Spiking Neural Network (SNN)

D) Radial Basis Function Network (RBFN)

Answer: C) Spiking Neural Network (SNN) (Correct)

12. What is the primary purpose of a "Gated Recurrent Unit" (GRU) in neural networks?

A) Introduce non-linearity to the model

B) Enhance the interpretability of the model

C) Prevent vanishing gradient problems in RNNs

D) Generate realistic images

Answer: C) Prevent vanishing gradient problems in RNNs (Correct)

13. What is the "Singular Value Decomposition" (SVD) commonly used for in machine learning?

A) Image compression

B) Time series prediction

C) Language translation

D) Speech synthesis

Answer: A) Image compression (Correct)

14. What is the term for the process of reducing the dimensionality of data while preserving its
important features?

A) Dimensionality Reduction

B) Feature Extraction

C) Data Smoothing

D) Data Augmentation

Answer: A) Dimensionality Reduction (Correct)

15. Which AI technique is inspired by the behavior of social insects like ants and bees and involves
decentralized decision-making?

A) Swarm Intelligence

B) Neural Evolution

C) Quantum Computing

D) Chaotic Dynamics

Answer: A) Swarm Intelligence (Correct)

16. What is the term for the ability of AI systems to understand and interpret human emotions
based on textual or verbal communication?

A) Emotional Intelligence

B) Sentiment Analysis

C) Empathy Analysis

D) Affective Computing

Answer: D) Affective Computing (Correct)

17. Which multinational corporation is known for its "BrainGate" project involving brain-computer
interfaces to restore communication?

A) Microsoft

B) Neuralink (Tesla)

C) IBM

D) Amazon

Answer: B) Neuralink (Tesla) (Correct)

18. What is the term for the mathematical representation of the connections between neurons in an
artificial neural network?

A) Synaptic Network

B) Connection Matrix

C) Weighted Graph

D) Neural Architecture

Answer: B) Connection Matrix (Correct)

19. What is the term for the process of analyzing and understanding the relationship between
various data points?

A) Data Exploration

B) Data Correlation

C) Data Association

D) Data Dependency

Answer: B) Data Correlation (Correct)

20. What is the term for the process of enhancing the quality and relevance of search results
through AI techniques?

A) Search Engine Optimization

B) Information Retrieval

C) Search Engine Enhancement

D) Contextual Ranking

Answer: B) Information Retrieval (Correct)

LEVEL 2 - AI Project Challenge

Terms & Conditions:

• Teams can choose any one among the two given problem statements.
• Time Limit: Teams have 45 minutes to complete the project challenge.
• Project Assessment: Evaluate based on predefined criteria.
• Results: Final scores will determine winners' advancement in the event.

1. Question:

Problem Statement: Given a dataset of toys and their features, build a program that groups toys
based on their similarities.

Teams should:

• Explain how the program groups toys together.

• Describe the features they used to decide on groups.
• Share how this could help organize things.

2. Question:

Problem Statement: Use a dataset of animals (cats & dogs) and their sounds to build a program that
matches animals to their sounds.

Teams should:

• Explain how the program connects animals with their sounds.

• Describe what helped them make the right matches.
• Share why knowing animal sounds can be fun.