ML Interview Questions

Describe the evolution of CNN architectures for image classification from AlexNet to modern models. What key innovations improved their performance over time?

Discuss modern approaches to implementing Optical Character Recognition (OCR) using deep learning models. How do these models address challenges such as varying fonts, languages, and image distortions?

Explain the differences between semantic, instance, and panoptic segmentation in computer vision. What are the challenges and recent advancements in each of these approaches?

Describe the pipeline of a facial recognition system, focusing on the stages from detection to identification, including any preprocessing steps, feature extraction methods, and classification techniques used.

Explain how you would handle class imbalance when working with image classification datasets. What are some techniques you can employ, and what are the potential benefits and drawbacks of each method?

Explain the architecture and working of Convolutional Neural Networks (CNNs) in detail. Discuss why they are particularly suited for image processing tasks and describe the advantages they have over traditional neural networks when dealing with image data.

Explain the role and functioning of convolutional layers in Convolutional Neural Networks (CNNs). How do they differ from fully connected layers, and why are they particularly suited for image processing tasks?