ML Interview Questions

Explain gradient boosting algorithms. How do they work, and what are the differences between XGBoost, LightGBM, and CatBoost?

What are the main approaches to feature selection in machine learning? Discuss the advantages and disadvantages of filter, wrapper, and embedded methods.

Explain the difference between supervised and unsupervised learning, and provide examples of algorithms used in each. Additionally, discuss the types of problems each is best suited to solve.

The concept of the 'curse of dimensionality' is often mentioned in the context of machine learning and data analysis. Can you explain what this term means and discuss its implications on model training and performance? Additionally, illustrate your explanation with an example of how adding dimensions can affect a k-nearest neighbors algorithm.

Explain how Principal Component Analysis (PCA) reduces dimensionality and discuss a scenario where applying PCA might improve a machine learning model's performance. What are some of the potential drawbacks of using PCA?

Can you explain the K-means clustering algorithm, including its step-by-step process, limitations, and practical applications? Additionally, when would K-means be the most appropriate choice for clustering data?

Can you explain the bias-variance tradeoff in machine learning? How does this tradeoff influence your choice of model complexity and its subsequent performance on unseen data?

Provide a comprehensive explanation of ensemble learning methods in machine learning. Compare and contrast bagging, boosting, stacking, and voting techniques. Explain the mathematical foundations, advantages, limitations, and real-world applications of each approach. When would you choose one ensemble method over another?

Can you describe how decision trees use information gain to decide which feature to split on at each node? How does this process contribute to creating an efficient and accurate decision tree model?

Imagine you are working on a binary classification task and your dataset is highly imbalanced. Explain how you would approach evaluating your model's performance. Discuss the limitations of accuracy in this scenario and which metrics might offer more insight into your model's performance.