Grokking Explained: How Neural Networks Suddenly 'Understand' Complex Logic
Feb 27 · 2 min read · Have you ever wondered why a neural network can struggle with a simple math problem for thousands of steps, only to suddenly 'get it' in a flash of insight? This phenomenon is known as Grokking, and it remains one of the most fascinating mysteries in...
Join discussionProbability Distributions in ML
Feb 17 · 6 min read · In Machine Learning, nothing is certain.When my model looks at a photo of a cat, it doesn't say "This is a Cat." It says: "I am 92% confident this is a cat, given the pointy ears and whiskers." To become a Machine Learning Engineer, I had to rewire m...
Join discussion
Essential Math Algorithms for Competitive Coding
Feb 16 · 7 min read · Binary Exponentiation : is a fast method to calculate powers like aⁿ , without multiplying a again and again n times. Instead of O(n) multiplications, it works in O(log n) time. Idea Behind Binary Exponentiation We use two simple observations : If po...
Join discussion
The Unified Theory of Matrices & Vector Spaces
Feb 16 · 9 min read · 1. Anatomy of a Linear System Before we draw a matrix, we have to understand where it comes from: Systems of Linear Equations. A generic linear system with m equations and n unknowns takes the form: $$\begin{align*} a_{11}x_1 + a_{12}x_2 + \dots + a_...
Join discussion
My Journey to Building a Neuro-Symbolic Math Agent
Feb 16 · 4 min read · I’m just an average developer who loves building ambitious systems from scratch. Currently, I am a sophomore at KL University, Hyderabad majoring in CSE. I love to procrastinate a lot and I thought maybe starting a blog might help me stay consistent ...
Join discussion