WorksheetsAI Arena - Round1
Total questions: 15
Worksheet time: 8mins
Aanya is working on a project where she needs to analyze a large dataset with many features. To simplify her analysis and improve the performance of her machine learning model, she decides to use a technique for dimensionality reduction. Which of the following algorithms should she choose?
Random Forest
PCA
Gradient Boosting
KNN
In supervised learning, the algorithm learns from:
Unlabeled Data
Labeled Data
Both A & B
None
The vanishing gradient problem is most common in:
CNNs
RNNs
SVMs
Decision Trees
True or False: NLP models convert text into numerical features using embeddings.
True
False
What does 'epoch' mean in training a neural network?
One complete pass through the entire dataset
One gradient update
A random batch
None
In NLP, 'stemming' and 'lemmatization' are used to:
Detect entities
Reduce words to their root form
Find sentiment
Tokenize
In a machine learning project, Arjun is training a neural network to classify images. He notices that the model performs well on the training data but poorly on the validation data. To improve the model's performance, he decides to implement dropout layers. Dropout layers are mainly used to:
Reduce bias
Prevent overfitting
Increase training speed
Normalize data
True or False: The softmax function outputs values between -1 and 1.
True
False
The loss function used in binary classification is usually:
Mean Squared Error
Cross Entropy
Hinge Loss
Log Loss
In NLP, which model architecture is based on self-attention?
LSTM
Transformer
CNN
RNN
True or False: Overfitting means the model performs well on training but poorly on unseen data.
True
False
What is 'tokenization' in NLP?
Breaking text into words or subwords
Converting numbers to text
Removing stopwords
Translating text
Gradient Descent helps to:
Minimize loss
Maximize accuracy
Find gradients
Split dataset
True or False: CNNs are mainly used for sequence prediction tasks.
True
False
Shaan is working on a machine learning project and needs to choose an activation function for his neural network. He is considering various options but wants to identify which of the following is NOT an activation function.
ReLU
Sigmoid
Softmax
Gradient
