Abhinav Anand·1d agoWeight Entropy Dynamics During CNN Training: Preliminary Evidence of Entropy Stabilization During…A preliminary experimental study using MNIST and Fashion-MNIST
Rusira Pinto·Nov 6Backpropagation, Step by Step: How a Neural Network Actually LearnsWorking through the chain rule by hand on a small network, one gradient at a time.
Aaryan Kakad·Nov 5exploring the back propagation paperthis is a raw and random exploration write-up, where I go through the back propagation paper and write what I feel in my own words.
Nandini Gundimeda·Nov 5Activation Functions in Deep LearningThis blog is explaining the fundamental role of activation functions in Neural Networks. A Neural Network is a machine learning model…
Piyas De·Nov 2Derivatives and Backpropagation: How AI Finds Who Was Responsible for the ErrorIn the previous article, we discussed gradient descent. Here is the link —…A response icon1A response icon1
InAI Simplified in Plain EnglishbyFrank Morales Aguilera·Oct 30Architectural Permanence: How Topo-CBP Muse-Glimmer-30B Overcomes Silicon FragilityFrank Morales Aguilera, BEng, MEng, SMIEEE
nachiket tanksale·Oct 30Day 4: Neural Network Backpropagation: What Is Actually Being Differentiated?Read Day 3 hereA response icon1A response icon1
Thakuriaditya·Oct 28From 2017 to Today: How the Transformer Quietly Rebuilt ItselfEvery modern LLM still looks like the 2017 transformer diagram from the outside, but four of its core parts have been replaced…
InData And BeyondbyDima Iakubovskyi·Oct 25Backpropagation 101: 40 Lines of NumPy That Learn XORA tiny network with no framework learns a problem no straight line can solve, and every gradient it computes matches a numerical check to…
Anes Bennani·Oct 22Target Propagation: Training Neural Networks Backwards Using “Goals” Instead of GradientsInstead of sending calculus errors down the network, what if each layer simply told the layer below what output it wanted to see?
