Dec 25, 2025 · 8 min read · From Zero to Foundation Models series is not necessarily aimed at absolute beginners. It is meant for people with some prior knowledge of AI, no matter how little it might be.The series is intended to act as a companion to readers: You have an inter...
Join discussion
Dec 25, 2025 · 19 min read · We have established that W is a measure of how important X is to our final decision. In the scenario we ran in module 1, I manually assigned the weights since we were only dealing with 2 inputs. Imagine we were dealing with hundreds of thousands of i...
Join discussion
Dec 17, 2025 · 6 min read · Introduction: Why Backpropagation Matters Backpropagation is the learning engine behind modern deep learning. Every time a neural network improves its predictions — from recognizing faces to generating text — the improvement comes from one mechanism:...
Join discussionNov 30, 2025 · 20 min read · The origins: Neural Networks and the human brain Neural networks were inspired by the network of neurons and synapses in the biological brain. A biological neuron receives input from its dendrites, performs some operation on the inputs, and then fire...
Join discussion
Sep 2, 2025 · 4 min read · Large language models (LLMs) such as GPT, LLaMA, and Falcon have billions of parameters. While these models are powerful, their size makes them challenging to run and adapt for specific use cases. Three key concepts help solve this problem: quantizat...
Join discussion
Jun 6, 2025 · 11 min read · On December 5, 2017, something remarkable occurred in a peaceful research laboratory that would forever alter our comprehension of artificial intelligence. An AI system named AlphaZero was provided with nothing but the fundamental rules of the old ga...
Join discussion
May 6, 2025 · 3 min read · In the world of artificial intelligence and optimization, differentiability is not just a mathematical nicety — it’s a critical enabler of nearly everything we train, tune, and learn. But why does smoothness matter so much? And what happens when our ...
Join discussion
Jan 24, 2025 · 28 min read · Neural networks have become a central pillar of modern machine learning, powering a broad range of technologies—from facial recognition systems on social media platforms to voice assistants in our smartphones. They draw inspiration from the biologica...
Join discussionJan 21, 2025 · 21 min read · In my last three blog posts, we explained about how to reduce the wrong but high confidence of neural network, how to identify dead neurons, and The Kaiming initialization and batch normalization. The core of these three blog post is, the correct ini...
Join discussion
Dec 30, 2024 · 9 min read · My last blog explained how to correctly initialize the last softmax layer of a neural network to reduce high but incorrect confidence when predicting certain classes. It also covered how to achieve uniform logits, a proper probability distribution, a...
AAbdul commented