An Overview of Machine Learning Model Training
ML training: Initialize model with random weights, forward pass makes predictions, calculate loss measuring error, backpropagation updates...
Welcome to the complete index of every article in our Terms Explained collection on When Notes Fly. This page lists every article in the section, organized alphabetically for easy reference. Each piece is researched, written by hand, and grounded in academic sources, professional practice, or empirical data. Whether you are diving into Terms Explained for the first time or returning to find a specific article, the index below gives you direct access to the full collection within Explainers.
If you are new to Terms Explained, we recommend starting with the foundational explainers and definitions before moving on to specific case studies, applied frameworks, and deeper analytical pieces. Articles are written for thoughtful readers who want substance over summary, with clear explanations of how ideas connect, where they come from, and why they matter. Use this index as a navigational map: skim the titles, read the short summaries, and click through to the pieces that draw your interest. Each article also links to related material so you can follow a thread of ideas across our entire Explainers library.
Most articles in this collection run between 1,500 and 3,000 words. We aim for the kind of explainer that holds up six months later: enough mechanism to be useful, enough nuance to be honest, and enough citation that you can verify the claims yourself. Where the research disagrees or the evidence is thin, we say so. Where a claim is well-established, we say that too. The goal is for you to leave with a working model you can apply, not a vibe you'll forget by Tuesday.
Bookmark this index — it gets fresh entries weekly. New articles are added at the top of the chronological feed and integrated into this alphabetical archive. If you can't find what you are looking for, try the broader Explainers archive for related ideas across all of Explainers, or browse our homepage for the latest writing.
ML training: Initialize model with random weights, forward pass makes predictions, calculate loss measuring error, backpropagation updates...
APIs define software communication contracts. REST APIs use endpoints like /users and HTTP methods like GET for read and POST for create operations.
Database indexes use B-tree structures maintaining sorted pointers to rows. Like book indexes, they enable fast lookups without scanning entire...
Encryption transforms plaintext into ciphertext using algorithms and keys. Intercepted data is useless without the key.
Load balancers distribute incoming requests across servers using algorithms like round robin for fairness and least connections for optimal routing.
Git stores snapshots as commits with complete file trees and metadata. Objects include blobs for content, trees for directories, commits for history.
Search engines crawl pages by following links, index content by extracting text and metadata, then rank results using algorithms and relevance...
DNS resolution: Browser checks cache, queries recursive resolver like Google DNS, resolver checks cache, then queries root nameservers to find IP...
Containers use Linux namespaces for isolated processes and cgroups for resource limits. Lightweight virtualization with separate filesystems and...
Transactions treat operations as single units—all succeed or all fail. ACID properties: Atomicity (all-or-nothing), Consistency, Isolation,...