An educational crypto, finance, and AI vocabulary puzzle about block metadata, invoice details, neural network training, and structured data formats.
Block Header Fields
Block headers contain compact metadata used to identify blocks and connect them to prior chain history.
Parent HashMerkle RootBlock TimestampBlock Number
- Parent Hash: A parent hash identifies the preceding block and links a block to its chain history.
- Merkle Root: A Merkle root is a single hash that commits to a collection of transactions organized in a Merkle tree.
- Block Timestamp: A block timestamp records a time associated with a block, subject to the rules of its blockchain.
- Block Number: A block number indicates the height or sequential position of a block in a blockchain.
Invoice Fields
Invoice fields identify a bill, state when payment is due, and describe the amounts charged.
Invoice NumberDue DateLine ItemSubtotal
- Invoice Number: An invoice number is an identifier assigned to a bill to distinguish it from other invoices.
- Due Date: A due date is the date by which payment of an invoice is expected.
- Line Item: A line item describes a specific good or service and its associated charge on an invoice.
- Subtotal: A subtotal is the sum of selected invoice charges before additional items such as taxes are applied.
Training Loop Steps
A training loop computes predictions and loss, derives gradients, and adjusts model parameters.
Forward PassLoss CalculationBackpropagationOptimizer Step
- Forward Pass: A forward pass applies a neural network to input data to compute its outputs.
- Loss Calculation: Loss calculation evaluates an objective that measures how well model outputs match the training goal.
- Backpropagation: Backpropagation applies the chain rule to compute gradients of a loss with respect to neural network parameters.
- Optimizer Step: An optimizer step uses gradients and an optimization rule to adjust trainable model parameters.
Data Serialization Formats
Serialization formats encode structured information so software systems can store it or exchange it.
JSONXMLYAMLProtocol Buffers
- JSON: JSON is a text format that represents structured data using objects, arrays, strings, numbers, booleans, and null.
- XML: XML is a markup format that represents structured data with elements, attributes, and text.
- YAML: YAML is a data serialization language commonly used for configuration and structured text documents.
- Protocol Buffers: Protocol Buffers is a serialization format that uses defined message schemas to encode structured data compactly.