Scribd
Upload
17 views3 pages

ML Interview Notes

The document provides an overview of Machine Learning (ML) and its subfields, including types of algorithms such as supervised, unsupervised, and reinforcement learning. It also covers concepts like Deep Learning, data preprocessing, feature engineering, and model evaluation, along with practical applications and future trends in ML. Additionally, it explains key terms and techniques like overfitting, regularization, and the bias-variance tradeoff.

Uploaded by

aniruddhasai720
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
17 views3 pages

ML Interview Notes

The document provides an overview of Machine Learning (ML) and its subfields, including types of algorithms such as supervised, unsupervised, and reinforcement learning. It also covers concepts like Deep Learning, data preprocessing, feature engineering, and model evaluation, along with practical applications and future trends in ML. Additionally, it explains key terms and techniques like overfitting, regularization, and the bias-variance tradeoff.

Uploaded by

aniruddhasai720
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

1. What is Machine Learning?

Machine Learning (ML) is a subfield of Artificial Intelligence (AI) that allows systems to learn from data
and make decisions or predictions without being explicitly programmed.
Example:​
Spam detection in Gmail that improves based on user behavior.

2. Types of Machine Learning Algorithms


●​ Supervised Learning: Trained on labeled data.
●​ Unsupervised Learning: Finds patterns in unlabeled data.
●​ Reinforcement Learning: Learns by interacting with an environment using rewards and
penalties.
Examples:
●​ Supervised: Loan approval prediction.
●​ Unsupervised: Customer segmentation.
●​ Reinforcement: Self-driving cars or game-playing bots.

3. What is Deep Learning?


Deep Learning is a subset of ML based on artificial neural networks with multiple layers, designed to
model complex patterns in large data.
Example:​
Facial recognition systems like Apple Face ID.

4. Layers in Deep Learning


●​ Input Layer: Accepts raw input.
●​ Hidden Layers: Perform calculations and feature extraction.
●​ Output Layer: Produces final prediction or result.
Example:​
In an image classifier, pixel values go to the input layer, then processed in hidden layers, and classified
in the output.

5. What is Backpropagation?
Backpropagation is a training algorithm that adjusts neural network weights by minimizing the error
using gradients calculated from the loss function.
Example:​
Used in handwriting recognition to reduce prediction error.

6. What is Data Preprocessing?


It’s the step of cleaning, transforming, and preparing raw data for machine learning models.
Example:​
Removing missing values, encoding categorical variables, scaling data.

7. What are LLMs (Large Language Models)?


LLMs are deep learning models trained on massive text data to understand and generate human
language.
Examples:​
ChatGPT, BERT, GPT-4 – used in virtual assistants, search engines, and content creation.

8. Real-time ML Applications on the Internet


The deployment of ML algorithms in online services and web applications to enhance user experience.
Examples:
●​ YouTube video recommendations
●​ Amazon product suggestions
●​ Fraud detection in online payments

9. Python Program to Sort a List Without Using Built-in Sort


def sort_list(lst):
for i in range(len(lst)):
for j in range(i+1, len(lst)):
if lst[i] > lst[j]:
lst[i], lst[j] = lst[j], lst[i]
return lst
print(sort_list([5, 2, 9, 1]))
10. Future Trends in ML for Internet
Emerging technologies and improvements shaping the future of machine learning on web-based
systems.
Examples:
●​ Federated Learning: ML on user devices for privacy
●​ Edge AI: Smart processing on IoT devices
●​ Explainable AI (XAI): Transparent decision-making in models

11. What is the difference between AI, ML, and DL?


AI (Artificial Intelligence): Broader concept of machines mimicking human intelligence.
ML (Machine Learning): Subset of AI that learns from data.
DL (Deep Learning): Subset of ML using neural networks.

12. What is Feature Engineering?


The process of selecting, modifying, or creating new input features from raw data to improve model
performance.
Example: Extracting "day of the week" from a date column in an e-commerce dataset.

13. What is Overfitting and Underfitting?


Overfitting: Model performs well on training data but poorly on unseen data.
Underfitting: Model performs poorly on both training and test data.
Solution:
Regularization, dropout, early stopping, more data.

14. What is Regularization in ML?


Technique to prevent overfitting by adding a penalty term to the loss function.
Types:
L1 (Lasso)
L2 (Ridge)

15. Explain the Bias-Variance Tradeoff.


Bias: Error due to simplistic model assumptions.
Variance: Error due to sensitivity to small fluctuations in training data.
Goal:
Achieve a balance for good generalization.

16. Difference Between Classification and Regression


Classification: Predict categories (e.g., spam/not spam)
Regression: Predict continuous values (e.g., house price)

17. What is a Confusion Matrix?


A table showing correct and incorrect predictions for classification models.
Terms:
True Positive, False Positive, True Negative, False Negative.

18. What is Precision, Recall, F1 Score?


Precision: TP / (TP + FP)
Recall: TP / (TP + FN)
F1 Score: Harmonic mean of precision and recall

19. Explain Random Forest Algorithm.


An ensemble method combining multiple decision trees using bagging.
Use:
Robust classification and regression on tabular data.

20. Difference Between Linear and Logistic Regression


Aspect​​ ​ Linear Regression​ ​ Logistic Regression
Output​​ ​ Continuous Value​ ​ Probability (0 to 1)
Function​ ​ Line Equation​ ​ ​ Sigmoid Function
Used For​ ​ Predicting numbers​ ​ Binary classification
21. What is Dropout in Deep Learning?
A regularization technique where neurons are randomly ignored during training to prevent overfitting.

22. Difference Between RNN and FFNN


Feature​ FFNN (Feedforward NN)​ ​ RNN (Recurrent NN)
Memory​ No memory​ ​ ​ ​ Has memory
Use Case​ Static data​ ​ ​ ​ Sequence data (e.g., time series, text)
Data Flow​ One direction​ ​ ​ ​ Loops over time steps

23. How is Machine Learning used on the Internet?


Recommendation engines (YouTube, Netflix)
Chatbots
Fraud detection
Predictive text
Ads personalization

24. What are Hyperparameters?


External configuration parameters set before training begins (e.g., learning rate, epochs, batch size).

25. What is Model Evaluation?


Techniques:
Cross-validation
Train/Test split
Metrics: Accuracy, Precision, Recall, RMSE, MAE

You might also like