🌱 Nexus Log

The Map of Becoming.

☀️ Today's Learnings

Parametric and Non-Parametric Learning

How much to assume about the world before observing it? The parametric/non-parametric distinction is an epistemic stance toward knowledge itself.

machinelearningpatternrecognitionbiasvariancetradeofftradeoffparametricgenerativemodelregularizationbayesianstatistics

Updated: 2026-01-06

📚 Recent History

29 past notes

Sat, Dec 27th

वीर भोग्य वसुन्धरा

संसार में अधिकार जन्म या भाग्य से नहीं, बल्कि कर्म, साहस और निरंतर प्रयास से मिलता है। जो आगे बढ़ते हैं, वही पृथ्वी के फल प्राप्त करते हैं।

philosophypsychologyvalourarmydarwinevolutionnietzschepowerhuman-nature

Sun, Dec 14th

The Turin Horse: A Philosophical Anecdote at the Intersection of Thought and Life

Anecdote of Nietzsche embracing the Horse suffering while carrying Load and the Master of the Horse goes on beating the Horse!

philosophymeaningbeliefperspectivepsychologynietzscheexistentialismanecdoteethicshuman-nature

Sat, Dec 13th

The Mind’s Search for Meaning in an Ambiguous World

Human brain has a deep-seated need to explain events, the drive to connect causes, effects, and meanings even where none exist. Our compulsion to find reasons both empowers and deceives us. What it might mean to meet reality without interpretation?

philosophymeaningpowerbeliefperspectivepropagandacriticalthinkingpsychology

Fri, Dec 12th

How Language Engineers Reality

An exploration of Alfred Korzybski's general semantics and its modern implications, revealing how words function as instruments of perception control. From corporate euphemisms to media framing, this piece exposes the Stimulus-Symbol-Response Loop that governs human behavior through linguistic manipulation.

philosophylanguagemeaningpowerlogicperspectivepropagandacriticalthinkingpatternrecognition

Thu, Dec 11th

The Data Bottleneck: How Modern ML Systems Fail and How to Build Ones That Dont

Data science has shifted from a compute-centric discipline to a data-intensive, systems-oriented one. It explains why traditional modeling skills are no longer enough and why production success depends on understanding data pipelines, reliability engineering, scalability constraints, distributed systems behavior, maintainability, and real-world fault tolerance. It reframes the role of the data scientist as a builder of resilient sociotechnical systems with an emphasis on operability, latency, robustness, and long-term evolution.

machinelearningsystemdesignstrategydatascience

Wed, Dec 10th

Auto-Scaling: The Hidden Engine Behind Modern Platforms and Real-Time Machine Learning

A deep dive into why auto-scaling is essential for platforms like Zomato, Swiggy, Uber, and Netflix to survive radical demand uncertainty. Explores how dynamic infrastructure enables reliability, cost-efficiency, and cutting-edge machine learning at massive scale.

machinelearningcloudcomputingdatascienceautoscalingdatascience

Tue, Dec 9th

The Electromagnetic Spectrum: A Universe in a Single Number

All electromagnetic phenomena from the hum of electrical circuits in your walls to the light you see, from radio waves to cosmic rays are the same thing, distinguished only by one number: frequency of oscillation.

physicselectromagnetismwavefeynman

Mon, Dec 8th

I am Dim and Confused while the People are Clever and Wise!

The certainty with which a system holds its maps inversely correlates with its contact with the territory. The more fixed your categories (good/evil, yes/no, fear/safety), the more energy you expend maintaining the fiction, and the less bandwidth remains for perceiving what is actually occurring.

philosophytaoperspectiveentropypsychologyevolution

Sun, Dec 7th

The Limits of Language and What Cannot Be Said

The limits of your Language are the limits of your Reality!

philosophylanguagemeaningpsychologypowerwittgensteinmathematicslogic

Sat, Dec 6th

The Picture Theory of Meaning

Most confusion in thinking comes from category errors, using factual language for non-factual domains, or vice versa. Stop seeking foundations where none exist. Philosophy's job isn't to build grand theories about the nature of reality, mind, or truth. It's to untangle confusions, to show you where you've mistaken grammar for metaphysics.

philosophywittgensteinpsychologyethicsproblem-framingmeaning

Fri, Dec 5th

The Expanding Frontier of Ignorance

Every piece of knowledge is merely an approximation to complete truth. All things must be learned only to be unlearned again, or corrected. The test of all knowledge is experiment, yet the source of knowledge requires imagination to guess at patterns, then experiment again to check the guess. Even a very small effect sometimes requires profound changes in our ideas.

feynmanphysicsperspectivesystemthinkingphilosophypatternrecognition

Thu, Dec 4th

Why Humans Blush, Laugh, Kiss, and Perform “Irrational” Acts?

Human involuntary behaviors like blushing, laughter, kissing, and even taboo acts are evolutionary truth-leak mechanisms designed to stabilize cooperation in a highly intelligent, deceptive species. They function as biological transparency systems that enforce honesty, synchronize groups, and maintain social cohesion across all levels of human interaction.

perspectiveemotionshuman-naturerelationshipsevolutionpsychologybehavior

Wed, Dec 3rd

Statistics as the Art of Transformation

Statistics is not merely the science devoted to organizing, summarizing, and ultimately making sense of such information. Statistics is the discipline that turns raw experience into structured understanding. Statistics answers the question: How do we convert the overwhelming chaos of the world into patterns we can understand, trust, and act upon?

statisticsperspectivesystemthinkingstatecraftphilosophy

Tue, Dec 2nd

Markov's Revenge: Chains of Consequence

What if the most powerful technologies in human history: atomic bombs, search engines, artificial intelligence were all accidents? What if they emerged not from grand visions, but from a Russian mathematician's fury at a religious rival? How Markov chains, invented in 1906 to refute a proof of God's existence, became the computational foundation of modernity. It's a story about intellectual warfare, unintended consequences, and the terrifying realization that we live inside abstractions we never chose and cannot escape.

probabilityartificialintelligencepatternrecognition

Mon, Dec 1st

The Art of Asking the Right Questions in Data Science

Data only answers the questions it’s capable of answering. Good analysis depends on asking clear, assumption-based questions that directly solve the real problem, otherwise even perfect models produce useless or misleading results.

systemthinkingproblem-framingdatascienceperspective

Sun, Nov 30th

Fundamental Physics behind Artificial Intelligence

Structural links between Artificial Intelligence and fundamental Physics

artificialintelligencedeeplearningmachinelearningphysicsperspectivequantumcomputing

Sat, Nov 29th

Logic vs Probability: The Dual Worlds of Software and Data Science

Philosophical difference between Data Scientist and Software Engineer

datascienceperspectiveproblem-framingsystemthinkingstrategyprobability

Fri, Nov 28th

Entropy, Surprise, and Coding Limits in Machine Learning

Shannon entropy as average surprise and minimum lossless code length, using dice and coding examples to link uncertainty, compressibility, and optimal coding. Covers uniform vs nonuniform distributions, Shannon’s noiseless coding theorem, maximum-entropy derivation via Lagrange multipliers, and connections to cross-entropy, KL divergence, decision trees, and data compression.

informationtheoryentropymachinelearningpatternrecognition

Thu, Nov 27th

Cambridge Analytica Scandal

A concise account of how Cambridge Analytica harvested Facebook data from 87 million users, built psychological profiles, and used precision-targeted political messaging during the 2016 U.S. elections, revealing the power of data-driven behavioral manipulation and Facebook’s role in enabling it.

datasciencestrategypsychologyperspectivehuman-natureemotionsethicspatternrecognitioninformation theory

Epistemic Warfare in Elections

A deep narrative excavation of Shivam Shankar Singh’s account of data-driven campaigning in How to Win an Indian Election.

datasciencestrategypsychologyperspectivehuman-natureethicspatternrecognitioninformation theoryelectionsdataanalysis

Han Feizi’s Law of Strategic Mismatch

Analysis of Han Feizi’s principle that destruction is inevitable when you attack order with chaos, legitimacy with deviance, or flexibility with rigidity.

statecraftstrategyhanfeizichinaperspectivepowerphilosophypsychology

Wed, Nov 26th

The Primacy of Simple Models

Start with a simple Model first, if it fails, you know you are in the wrong direction without going too far. So you can change the direction and route.

datasciencestrategyproblem-framingperspectivemachinelearningsystemthinking

Tue, Nov 25th

Information Theory: The Architecture of Uncertainty and Human Power

A detailed introduction to Information Theory with its application in almost whole of the Universe.

entropypatternrecognitionmachinelearningperspectivestrategyproblem-framingphilosophypsychologyhuman-natureentropyinformationtheory

The OKCupid Experiment: When Data Met Desire

How Chris McKinlay used Web Scraping and got Lucky.

ethicsdatasciencestrategyperspectivesystemthinkingwebscrapingpatternrecognition

Mon, Nov 24th

Generative and Discriminative Models

How Generative and Discriminative Models work.

machinelearningpatternrecognitiongenerativemodeldiscriminativemodel

Data-Centric, Not Data-Driven

Why Data-Centric is more appropriate term compared to Data-Driven.

datasciencestrategyproblem-framingperspective

The Symphony of Human Change

An exploration of human emotional volatility and the unconditional acceptance that makes love possible.

human-natureemotionsphilosophyloverelationshipspsychology

Sun, Nov 23rd

Exploration - Exploitation Curve in Reinforcement Learning

An examination of the fundamental tension in reinforcement learning between exploring new possibilities and exploiting known rewards.

machinelearningpatternrecognitionreinforcementlearningexploration-exploitationcurve

Sat, Nov 1st

Pattern Recognition and Machine Learning

A concise exploration of pattern recognition as the computational evolution of humanity's timeless quest to find order in data and generalize from examples.

machinelearningpatternrecognition