20 Questions
What type of learning uses labeled data with a specific goal?
Supervised Learning
In machine learning, what is used for predicting a categorical variable like 'cat' or 'dog'?
Logistic Regression
What is the step in data preparation that involves creating features that contribute to a prediction?
Feature engineering
Which type of layer in an Artificial Neural Network (ANN) sums each input multiplied by a weight?
Hidden layer
What should you do if the loss function oscillates during training?
Lower the learning rate
Which activation function converts the input to a value between 0 and 1 for binary class classification?
Sigmoid
What is used to calculate the difference between the NN prediction and the actual result?
Mean square error (MSE)
What does precision measure in the context of machine learning?
How many selected items are relevant
Which option provides the highest level of control and parameter tuning in AI model training?
Custom training
Which layer in TensorFlow provides building blocks to create machine learning model like neural network layers and evaluation metrics?
TF model libraries
What type of learning uses unlabeled data for pattern recognition?
Unsupervised learning
Which technique is used for predicting a numeric value like forecasting sales in machine learning?
Linear regression
What step in data preparation involves creating features that contribute to a prediction?
Feature engineering
Which layer of an Artificial Neural Network (ANN) has the weighted sum of hidden layers?
Output layer
What should be done to address a situation where the model doesn't fit well or converge?
Increase the number of epochs
Which function is responsible for converting a linear network to a non-linear network?
ReLU (rectified linear unit)
In the context of machine learning, what does gradient descent aim to find?
Minimum cost function
Which activation function is suitable for multi-class classification where the output is a probability distribution?
Softmax function
What should be the initial approach to batch size when training a model?
Start with a higher batch size and decrease until it is no longer stable
Which TensorFlow abstraction layer provides basic and numeric processing functions?
Core TensorFlow (C++, Python)
Test your knowledge of the data to AI workflow, including data ingestion, processing, storage, analysis, visualization, and machine learning basics such as supervised learning and classification. Explore concepts like PubSub, Dataflow, Dataproc, Cloud data fusion, BigQuery, Looker, VertexAI, autoML, and generative AI studio.
Make Your Own Quizzes and Flashcards
Convert your notes into interactive study material.
Get started for free
