Jul 17, 2026

Onchain Identity, Monetary Policy, Training Pipelines, and Decentralized Compute

A daily crypto, finance, and AI vocabulary puzzle covering identity primitives, central-bank tools, model training, and distributed compute.

Onchain Identity

These words describe identifiers and credentials that can connect accounts with verifiable claims.

Wallet AddressDecentralized IdentifierAttestationSoulbound Token
  • Wallet Address: A wallet address is a public identifier used to receive assets or interact with applications on a blockchain.
  • Decentralized Identifier: A decentralized identifier, or DID, is an identifier designed to be created and controlled without relying on one central registry.
  • Attestation: An attestation is a verifiable statement that an issuer makes about an account, credential, or event.
  • Soulbound Token: A soulbound token is a non-transferable token used to represent a credential, membership, or reputation signal.

Monetary Policy

These words explain policy rates, balance-sheet actions, and inflation-adjusted returns.

Policy RateOpen Market OperationsQuantitative TighteningReal Yield
  • Policy Rate: A policy rate is a benchmark interest rate a central bank sets to influence borrowing costs and economic activity.
  • Open Market Operations: Open market operations are purchases or sales of securities used by a central bank to manage liquidity and short-term rates.
  • Quantitative Tightening: Quantitative tightening reduces a central bank's asset holdings, which can withdraw liquidity from the financial system.
  • Real Yield: Real yield measures an investment's return after accounting for expected or realized inflation.

Training Pipelines

These words describe training data, optimization steps, and saved model states.

Training CorpusMini-BatchGradient DescentCheckpoint
  • Training Corpus: A training corpus is the collection of text, images, audio, or other examples used to train a model.
  • Mini-Batch: A mini-batch is a subset of training examples processed together during one optimization step.
  • Gradient Descent: Gradient descent updates model parameters in a direction intended to reduce the training loss.
  • Checkpoint: A checkpoint stores model parameters and related training state so work can be resumed or evaluated later.

Decentralized Compute

These words describe marketplaces, providers, routing, and verification for distributed AI compute.

Compute MarketplaceCompute ProviderInference NetworkVerifiable Compute
  • Compute Marketplace: A compute marketplace connects buyers of processing capacity with infrastructure providers that have available resources.
  • Compute Provider: A compute provider supplies GPUs, CPUs, storage, or networking resources to run workloads.
  • Inference Network: An inference network routes model requests to distributed workers that produce outputs.
  • Verifiable Compute: Verifiable compute uses proofs or validation methods to help others check that a computation was performed correctly.