InThe Quantastic JournalbyHarish K·Oct 6The Matrix Is Mostly Empty: An Intuitive (and Hands-On) Guide to Matrix FactorizationImagine a spreadsheet where the rows are users and the columns are movies. Each cell contains a rating from 1 to 5, but most of the cells…A response icon2A response icon2
Kaveesha Senarathne·Oct 5Number Theory in Computing: Algorithms, Cryptography, and Real-World ApplicationsNumber theory is a part of pure mathematics. Numbers can be classified into different types such as real numbers, rational numbers…
MR ROBOT·Sep 15The Transformer’s Footprints IIပြီးခဲ့တဲ့ post မှာဆိုရင် hidden state တွေအကြောင်းကို ရှင်းပြခဲ့ပြီးဖြစ်တဲ့အတွက် ဒီနေ့မှာတော့ Transformer model တွေကို အသုံးမပြုခင် AI…
Patrick CHER·Sep 8The Simple Recommender Wins, Except for One Kind of ReaderFour recommender tiers, 1.15M Book-Crossing ratings, scored separately for readers by reading history depth. The simplest personalized…
Ravi Krishna·Jun 6Cutting AI Costs Without Cutting Quality: A Two-Stage Routing ArchitectureWhy sending every query to your best model is the quiet source of runaway AI spend—and a simple, learnable fix.
Pallabi S Roy·May 13Machine Learning With Manya: Unlocking Netflix’s Mind-Reading CodeImagine you have a box of toys. You don’t know which toys your friends like. But you want to give each friend the toy they’ll love the…
Sudhanva Savyasachi S·Mar 28Matrix Factorisation Part-1The most important topic of classical Recommender systems is matrix factorisation.
Krishan Bansal·Feb 18The Beautiful Math Behind PCA: Part2 : Singular Value DecompositionA complete walkthrough of SVD from scratch — and how it makes PCA faster, cleaner, and more general.
Uri Kialy·Dec 17, 2025Decomposing MLP Activations into Interpretable Features via Semi-Nonnegative Matrix FactorizationTo those who are not new to mechanistic interpretability- describing features has been a main domain in this area in past years. In the…
