ML Interview Questions

What is agentic AI, and how does it differ from traditional AI in terms of function and design principles?

How Large Language Models (LLMs) handle out-of-vocabulary (OOV) words or tokens?

How do you reduce inference cost for LLMs?

Can you explain how Large Language Models (LLMs) are typically trained? What are the key components and phases involved in their training process?

Explain the difference between on-policy and off-policy reinforcement learning methods. How do these approaches impact the learning process and what are some examples of algorithms that use each method?

What is the credit assignment problem in Reinforcement Learning, and what strategies can be employed to effectively address it?

Explain how Q-learning works, its theoretical foundations, and list some common limitations. Additionally, provide practical examples where Q-learning can be effectively applied.

Explain the Policy Gradient Theorem and describe how the REINFORCE algorithm implements this concept in Reinforcement Learning.

Compare model-based and model-free reinforcement learning approaches, focusing on their theoretical differences, practical applications, and the trade-offs involved in choosing one over the other.

Explain the Proximal Policy Optimization (PPO) algorithm and discuss why it is considered more stable compared to traditional policy gradient methods.