Deep Learning Fundamentals
Perceptrons, backpropagation, activation functions, loss functions, optimizers, batch size, epochs
1What is a perceptron in the context of neural networks?
What is a perceptron in the context of neural networks?
Answer
A perceptron is the basic unit of a neural network, inspired by biological neurons. It takes multiple inputs, multiplies them by weights, sums everything with a bias, then applies an activation function to produce an output. The simple perceptron can only solve linearly separable problems, which led to the development of multilayer networks.
2What is the main limitation of the simple (single-layer) perceptron?
What is the main limitation of the simple (single-layer) perceptron?
Answer
The simple perceptron can only solve linearly separable problems, meaning problems where classes can be separated by a straight line (or hyperplane in higher dimensions). This limitation, demonstrated by Minsky and Papert in 1969 with the XOR problem, temporarily slowed neural network research until multilayer perceptrons were introduced.
3What is the role of the activation function in a neural network?
What is the role of the activation function in a neural network?
Answer
The activation function introduces non-linearity into the network, allowing it to learn complex relationships between inputs and outputs. Without a non-linear activation function, even a multi-layer network would behave as a simple linear transformation. Common functions include ReLU, sigmoid, and tanh, each with specific properties depending on the use case.
Which activation function is most commonly used in hidden layers of modern networks?
When should the softmax activation function be used in a neural network?
+21 interview questions
Other Data Science & ML interview topics
Python Basics
Python Object-Oriented Programming
Python Data Structures
Git Fundamentals
SQL Basics
NumPy Fundamentals
Pandas Basics
Jupyter & Google Colab
SQL Joins & Advanced Queries
Advanced Pandas
Visualization with Matplotlib & Seaborn
Interactive Visualizations with Plotly
Descriptive Statistics
Inferential Statistics
Web Scraping
BigQuery & Cloud Data
Feature Engineering
Supervised ML: Regression
Supervised ML: Classification
Decision Trees & Ensembles
Unsupervised ML
ML Pipelines & Validation
Time Series & Forecasting
TensorFlow & Keras
CNN & Image Classification
RNN & Sequences
Transformers & Attention
NLP & Hugging Face
GenAI & LangChain
MLOps & Deployment
Master Data Science & ML for your next interview
Access all questions, flashcards, technical tests, code review exercises and interview simulators.
Start for free