--- title: "#machine-learning articles | Hashnode" description: "Machine Learning is a subfield of Computer Science that evolved from the study of pattern recognition and computational learning theory in artificial intelli..." url: "https://hashnode.com/tag/machine-learning" --- Tag # #machine-learning 20,398 posts · 47,751 followers [Vashni Roy](https://hashnode.com/@vashniroy) · [4h ago](https://hashnode.com/posts/why-0-5-isn-t-always-the-right-classification-threshold/6af2ffa99538fc7c0f2ddfef) ### [Why 0.5 Isn't Always the Right Classification Threshold](https://vashniroy.hashnode.dev/why-0-5-isn-t-always-the-right-classification-threshold?utm_source=hashnode&utm_medium=feed) I changed the threshold of the same churn model from 0.3 to 0.7. The model stayed the same, but its predictions changed quite a lot. 3 min read · [vashniroy.hashnode.dev](https://vashniroy.hashnode.dev/why-0-5-isn-t-always-the-right-classification-threshold?utm_source=hashnode&utm_medium=feed) [Nagappan S](https://hashnode.com/@imkarthik03) · [5h ago](https://hashnode.com/posts/semantic-caching-skipping-the-model-call-when-you-ve-already-answered-the-question/6af2f1dcd29ff15e200dc8c5) ### [Semantic Caching: Skipping the Model Call When You've Already Answered the Question](https://nagspidey.hashnode.dev/semantic-caching-skipping-the-model-call-when-you-ve-already-answered-the-question?utm_source=hashnode&utm_medium=feed) Semantic Caching 7 min read · [nagspidey.hashnode.dev](https://nagspidey.hashnode.dev/semantic-caching-skipping-the-model-call-when-you-ve-already-answered-the-question?utm_source=hashnode&utm_medium=feed) [Siddhesh Sawant](https://hashnode.com/@siddheshsawant) · [14m ago](https://hashnode.com/posts/logistic-regression-from-first-principles/6af340fb4f49d259cfecfa43) ### [Logistic Regression from First Principles: Log-Odds, Likelihood and the Decision Threshold](https://siddheshsawant.hashnode.dev/logistic-regression-from-first-principles?utm_source=hashnode&utm_medium=feed) Logistic regression derived step by step: sigmoid and log-odds, log loss and its gradient, gradient descent and Newton's method, and the decision threshold. 53 min read · [siddheshsawant.hashnode.dev](https://siddheshsawant.hashnode.dev/logistic-regression-from-first-principles?utm_source=hashnode&utm_medium=feed) [Siddhesh Sawant](https://hashnode.com/@siddheshsawant) · [25m ago](https://hashnode.com/posts/linear-regression/6af33e7475a9ec13b893f93b) ### [Linear Regression from First Principles: Least Squares, Geometry and Gradient Descent](https://siddheshsawant.hashnode.dev/linear-regression?utm_source=hashnode&utm_medium=feed) Fitting a straight line is the oldest trick in machine learning and still the one most worth understanding completely. This article builds linear regression from a five-point example to the normal equations, shows why least squares is a projection and a maximum-likelihood estimate at the same time, trains it with gradient descent you can steer, and then puts it to work on a real dataset to see where it holds up and where it breaks. 48 min read · [siddheshsawant.hashnode.dev](https://siddheshsawant.hashnode.dev/linear-regression?utm_source=hashnode&utm_medium=feed) [Embrel Grim](https://hashnode.com/@embrelgrim) · [8h ago](https://hashnode.com/posts/an-attempt-at-hypernetworks-on-an-rx-6600/6af2cc3514a8a12b8ac6b5bd) ### [an attempt at hypernetworks on an RX 6600](https://grimoire.hashnode.dev/an-attempt-at-hypernetworks-on-an-rx-6600?utm_source=hashnode&utm_medium=feed) Where the Idea Came From I realized that my hardware couldn't run anything decent, so I decided to play it safe—buy proper components? Of course not—build my own architecture! Full of confidence, I sa 10 min read · [grimoire.hashnode.dev](https://grimoire.hashnode.dev/an-attempt-at-hypernetworks-on-an-rx-6600?utm_source=hashnode&utm_medium=feed) [wanqiuxu](https://hashnode.com/@wanqiuxu) · [1h ago](https://hashnode.com/posts/why-language-models-invent-academic-citations-that-look-real/6af329e21fbc9ddd79c36359) ### [Why language models invent academic citations that look real](https://crushsci.hashnode.dev/why-language-models-invent-academic-citations-that-look-real?utm_source=hashnode&utm_medium=feed) Ask a language model for a short literature review, or for "a few key papers," and you may get a list that looks professionally done: surnames, years, journal names, tidy titles. Then you search for o 4 min read · [crushsci.hashnode.dev](https://crushsci.hashnode.dev/why-language-models-invent-academic-citations-that-look-real?utm_source=hashnode&utm_medium=feed) [Yong Yu](https://hashnode.com/@yongbo) · [1h ago](https://hashnode.com/posts/muse-ember-at-0-10-0-50-default-is-max-the-200k-billing-cliff-and-when-to-pair-it-with-wonder/6af3294965dfe6eeccc46569) ### [Muse Ember at $0.10/$0.50: Default Is Max, the 200K Billing Cliff, and When to Pair It With Wonder](https://yongboyu.hashnode.dev/muse-ember-at-0-10-0-50-default-is-max-the-200k-billing-cliff-and-when-to-pair-it-with-wonder?utm_source=hashnode&utm_medium=feed) Meta Muse Ember is on the API today as muse-ember-1.0: $0.10/$0.004/$0.50 under 200K input, 3× rates on the whole request past that line, default reasoning is max, AA Index ~36. No chat call on this box. 10 min read · [yongboyu.hashnode.dev](https://yongboyu.hashnode.dev/muse-ember-at-0-10-0-50-default-is-max-the-200k-billing-cliff-and-when-to-pair-it-with-wonder?utm_source=hashnode&utm_medium=feed) [CaraComp](https://hashnode.com/@CaraComp) · [2h ago](https://hashnode.com/posts/deepfake-election-taiwan-warns-fakes-can-lie-about-voting/6af31aa012d97dbc2750368c) ### [Deepfake Election: Taiwan Warns Fakes Can Lie About Voting](https://caracomp.hashnode.dev/deepfake-election-taiwan-warns-fakes-can-lie-about-voting?utm_source=hashnode&utm_medium=feed) Analyzing the technical fallout of synthetic media in Taiwan's elections highlights a critical shift in information integrity: malicious actors no longer need to construct elaborate scandal narratives 3 min read · [caracomp.hashnode.dev](https://caracomp.hashnode.dev/deepfake-election-taiwan-warns-fakes-can-lie-about-voting?utm_source=hashnode&utm_medium=feed) [jovin george](https://hashnode.com/@softreviewed) · [3h ago](https://hashnode.com/posts/claude-fable-5-2-vs-mythos-5-2-speed-benchmarks-real-costs/6af31a2ef3c7ce0ae2ccbb7a) ### [Claude Fable 5.2 vs Mythos 5.2: Speed, Benchmarks & Real Costs](https://softreviewed.hashnode.dev/claude-fable-5-2-vs-mythos-5-2-speed-benchmarks-real-costs?utm_source=hashnode&utm_medium=feed) Anthropic's frontier model deep dive: $20 Pro plan trial, FrontierCode 2.0 hack, and silent fallbacks. 6 min read · [softreviewed.hashnode.dev](https://softreviewed.hashnode.dev/claude-fable-5-2-vs-mythos-5-2-speed-benchmarks-real-costs?utm_source=hashnode&utm_medium=feed) [mihalykrich](https://hashnode.com/@mkrich) · [7h ago](https://hashnode.com/posts/building-krxy-network-outage-qwen35-model-library-part-8/6af2d93c1d0be502ac51a1cb) ### [Building KRXY — Part 8: When the Network Slows Down — and the Model Library Grows](https://krxy.hashnode.dev/building-krxy-network-outage-qwen35-model-library-part-8?utm_source=hashnode&utm_medium=feed) 1–2 November 2026: two-Pi learning through bandwidth limits and a two-minute outage, followed by the first Qwen3.5 2B and 4B installations. 13 min read · [krxy.hashnode.dev](https://krxy.hashnode.dev/building-krxy-network-outage-qwen35-model-library-part-8?utm_source=hashnode&utm_medium=feed)