James Stewart's Calculus Chapter 3 — a concise summary of absolute extrema, the Mean Value Theorem, derivatives and the shape of graphs, optimization, Newton's Method, and antiderivatives.
James Stewart's Calculus, Chapter 2 — a concise summary covering derivatives and rates of change, the definition of the derivative, basic differentiation rules, the chain rule, implicit differentiation, related rates, and linear approximation.
James Stewart's Calculus Chapter 1 — a concise summary covering the definition and types of functions, the concept of limits and their computation laws, continuity, and limits involving infinity.
A collection of concept exercises applying differentiation, loss function optimization, gradient descent, and backpropagation to real-world machine learning scenarios. Each problem is grounded in situations you commonly encounter when training models in practice.
Hands-on coding assignments implementing numerical differentiation, gradient descent, perceptron backpropagation, and Newton's method from scratch with NumPy. Each problem includes step-by-step hints and complete solution code.
A deep dive into the principles of gradient descent and learning rate, perceptron regression and classification, backpropagation, and Newton's method with the Hessian — the core concepts of neural network optimization.
A deep dive into the core of machine learning optimization: minimizing loss functions, differentiating squared and log loss, understanding partial derivatives and gradients, and using the gradient to find minima.
Build an intuitive understanding of derivatives — the engine behind ML optimization — and master the differentiation rules for constants, polynomials, exponentials, logarithms, and trig functions, plus scalar multiplication, sum, product, and chain rules.