How do federated learning approaches differ from traditional machine learning in terms of data handling?
- Federated learning doesn't use data
- Federated learning relies on centralized data storage
- Federated learning trains models on decentralized data
- Traditional machine learning trains models on a single dataset
Federated learning trains machine learning models on decentralized data sources without transferring them to a central server. This approach is privacy-preserving and efficient. In contrast, traditional machine learning typically trains models on a single, centralized dataset, which may raise data privacy concerns.
For graph processing in a distributed environment, Apache Spark provides the _______ library.
- GraphX
- HBase
- Pig
- Storm
Apache Spark provides the "GraphX" library for graph processing in a distributed environment. GraphX is a part of the Spark ecosystem and is used for graph analytics and computation. It's a powerful tool for analyzing graph data.
In computer vision, what process involves converting an image into an array of pixel values?
- Segmentation
- Feature Extraction
- Pre-processing
- Quantization
Pre-processing in computer vision typically includes steps like resizing, filtering, and transforming an image. It's during this phase that an image is converted into an array of pixel values, making it ready for subsequent analysis and feature extraction.
Which of the following is not typically a layer in a CNN?
- Convolutional Layer
- Fully Connected Layer
- Recurrent Layer
- Pooling Layer
Recurrent Layers are not typically used in Convolutional Neural Networks. They are more common in Recurrent Neural Networks (RNNs) and are used for sequential data processing, unlike CNNs, which are designed for grid-like data.
In EDA, which method can help in understanding how a single variable is distributed across various categories or groups?
- Histogram
- Box Plot
- Scatter Plot
- Bar Plot
A bar plot is used to visualize the distribution of a single variable across different categories or groups. It displays the data in rectangular bars, making it easy to compare and understand how the variable is distributed among the categories. Commonly used in Exploratory Data Analysis (EDA).
You're working with a dataset containing sales data from various regions. You want to identify sales patterns, seasonal trends, and anomalies. Which EDA techniques and visualization tools would be best suited for this?
- Scatter plots and t-SNE
- Box plots and bar charts
- Time series plots and heatmaps
- Histograms and parallel coordinates
For exploring sales patterns and seasonal trends, time series plots and heatmaps are excellent choices. Time series plots can reveal trends over time, and heatmaps can show correlations between different regions and sales data, helping identify anomalies and patterns.
Which method in transfer learning involves freezing the earlier layers of a pre-trained model and only training the latter layers for the new task?
- Fine-tuning
- Knowledge Transfer
- Feature Extraction
- Weight Sharing
The method in transfer learning that involves freezing the earlier layers of a pre-trained model and only training the latter layers for the new task is known as fine-tuning. Fine-tuning allows the model to retain the knowledge from the source task while adapting its later layers for the specific requirements of the target task. This approach is common in transfer learning scenarios.
While working with a dataset about car sales, you discover that the "Brand" column has many brands with very low frequency. To avoid having too many sparse categories, which technique can you apply to the "Brand" column?
- One-Hot Encoding
- Label Encoding
- Brand grouping based on frequency
- Principal Component Analysis (PCA)
To handle low-frequency categories in the "Brand" column, you can group the brands based on their frequency. This reduces the number of sparse categories and can improve model performance. You can also consider techniques like label encoding or one-hot encoding, but they might not be ideal for low-frequency categories. PCA is used for dimensionality reduction and not for handling categorical variables.
A neural network without any hidden layers is typically referred to as a _______.
- Deep Neural Network
- Shallow Neural Network
- Multilayer
- Perceptron
A neural network without any hidden layers is often referred to as a "Perceptron." It consists of only the input and output layers, and it's the simplest form of a neural network.
_________ is a popular open-source framework used for real-time processing and analytics of large streams of data.
- Hadoop
- Spark
- Hive
- Kafka
Apache Spark is a widely used open-source framework for real-time processing and analytics of large streams of data. It provides powerful tools for data processing, machine learning, and more, making it a popular choice in the field of big data and data science.