Click a topic or expand the play button for filtered posts, or scroll to browse all.
An in-depth exploration of surrogate loss functions, their convexity properties, and Fisher consistency in binary classification, explaining the theory behind popular losses like hinge, exponential, logistic, and truncated quadratic.
A detailed yet succinct overview of convex optimization: from identifying convex functions and preserving convexity to formulating standard convex optimization problems, applying KKT conditions, and exploring popular convex loss functions and common convex functions.
A detailed yet succinct overview of convex optimization: from identifying convex functions and preserving convexity to formulating standard convex optimization problems, applying KKT conditions, and exploring popular convex loss functions and common convex functions.
A detailed yet succinct overview of convex optimization: from identifying convex functions and preserving convexity to formulating standard convex optimization problems, applying KKT conditions, and exploring popular convex loss functions and common convex functions.
An in-depth look at the support vector machine, including distance calculations, margin definitions, hard/soft margin formulations, and the kernel trick for nonlinear classification.
A reflective narrative on continuously creating purpose through mindful actions and self-dialogue.
Optimality of Conditional Expectation under MSE and Its Implications for the Bias–Variance Tradeoff
A concise extension covering the model setup, dimensions, derivations, assumptions, and real-world applications of multivariate (multinomial) logistic regression.
A concise overview of logistic regression, including MLE under the frequentist paradigm and Bayesian inference perspectives, with a focus on assumptions and model diagnostics.
A comprehensive overview of Bayesian linear regression, including the standard approach, Bayesian Ridge, and Bayesian Lasso with detailed derivations.
A concise overview of linear regression approaches, including OLS, Ridge, Lasso, and Elastic Net.
A concise overview of key vector and matrix derivatives used in optimization and machine learning, along with useful resources such as the Matrix Cookbook.
A concise overview of key matrix properties, including inverses, determinants, and traces.
A comprehensive template showcasing beautifully styled mathematical descriptions with different fonts and a collection of KaTeX-rendered math expressions, including inline math, display equations, matrices, piecewise functions, and more.
An exploration into modern living, questioning if we are truly living or just surviving amidst societal pressures.