Deep Learning — Prelims Questions

Which of the following statements about Deep Learning (DL) is/are correct? 1. DL models typically require large datasets for optimal performance. 2. Backpropagation is a core algorithm used for training deep neural networks. 3. Convolutional Neural Networks (CNNs) are primarily used for processing sequential data like text. 4. Transformer architecture relies on recurrent connections to process information.

Consider the following applications: 1. Early detection of diabetic retinopathy from retinal scans. 2. Predictive maintenance in industrial machinery. 3. Real-time translation of spoken languages. 4. Crop yield prediction based on satellite imagery. Which of the above are prominent applications of Deep Learning in India, particularly in line with the 'AI for All' vision?

Which of the following is NOT a primary ethical concern associated with the widespread deployment of Deep Learning models?

Which of the following Deep Learning models is primarily known for its 'bidirectional' training approach, allowing it to understand context from both left and right of a word in a sentence?

Consider the following statements regarding India's approach to Artificial Intelligence: 1. The 'AI for All' vision emphasizes inclusive growth and addressing societal challenges. 2. The National AI Portal (indiaai.gov.in) is a joint initiative of NITI Aayog and NASSCOM. 3. The Centre for Artificial Intelligence & Robotics (CAIR) is focused on AI applications for defence. Which of the statements given above is/are correct?