Explain key ML/stats concepts
Company: C3 AI
Role: Data Scientist
Category: Machine Learning
Difficulty: medium
Interview Round: Onsite
# Explain key ML/stats concepts
You are taking an ML/Stats screening with conceptual multiple-choice questions. Answer the following:
1. **CNN vs. RNN**
- What kinds of input structure does each model assume?
- Give 1–2 examples of tasks where CNNs are typically preferred and tasks where RNNs (or sequence models) are preferred.
2. **Gini impurity (decision trees)**
- Define Gini impurity for a node with class probabilities \(p_1,\dots,p_K\).
- Compute the Gini impurity for a binary node with \(p=0.8\) and \(1-p=0.2\).
3. **Entropy vs. Gini for split criteria**
- Define entropy for a node and compare it to Gini impurity.
- Explain how they differ in sensitivity and whether they usually produce meaningfully different trees in practice.
4. **Multicollinearity (linear/logistic regression)**
- What is multicollinearity and why is it a problem?
- Name at least two ways to detect it and two ways to mitigate it.
5. **Pearson correlation**
- Define Pearson correlation and list key assumptions/limitations.
- Explain why correlation \(\neq\) causation and give one example of confounding.
### Clarifying Questions to Ask
- Clarify the task, data shape, labels, constraints, and evaluation metric.
- State assumptions behind the math or modeling technique you choose.
- Connect theory to practical training, debugging, and deployment implications.
### What a Strong Answer Covers
- Correct definitions and formulas where the prompt requires them.
- A practical explanation of how the method behaves on real data.
- Trade-offs, failure modes, diagnostics, and mitigation strategies.
- Evaluation choices that match the product or modeling objective.
### Follow-up Questions
- How would noisy labels, class imbalance, or distribution shift affect the answer?
- What would you monitor after deployment?
- Which baseline would you compare against first?
Overview: Explain key ML/stats concepts evaluates core ML concepts, assumptions, math intuition, training/evaluation trade-offs, and practical failure modes in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.