Skip to content

MATH 105: Classical Probability and Statistics
Course Code MATH 105
Course Name Classical Probability and Statistics
Department Mathematics
Semester Offered Odd (Term 1)
Tuition Hours 10 hours
Course Level Foundational (Primer)
Pre-requisite -
Co-requisite MATH 101: Linear Algebra for Machine Learning, MATH 102: Calculus for Machine Learning
Course Objective This course exists for one reason: to make sure students are not confused when data starts talking back.

In Term 1, students will build AI systems, analyze user behavior, and make decisions based on data. Words like probability, distribution, correlation, and confidence will show up everywhere. Most students nod along without really understanding them. That is a problem.

This is a short, sharp introduction to the language of uncertainty. No depth for the sake of it. No long derivations. Just enough clarity so that when students see metrics, experiments, or model outputs, they know what they mean and when not to trust them.
Course Philosophy This course emphasizes
  • Clarity over completeness
  • Interpretation over calculation
  • Real-world meaning over mathematical rigor
The goal is not to train statisticians. The goal is to ensure students can reason about uncertainty, data variation, and decisions without getting lost in formulas.
Course Learning Outcomes Upon successful completion of this course, students will be able to:
  • Understand probability as a measure of uncertainty in real-world events.
  • Interpret basic statistical measures like mean, variance, and standard deviation.
  • Understand the idea of distributions, especially the normal distribution.
  • Interpret correlation correctly, and avoid common traps like assuming causation.
  • Understand confidence intervals as ranges of uncertainty, not guarantees.
  • Interpret p-values at a high level, especially in experiments and A/B testing.
  • Make basic data-driven decisions in their AI product without being misled by noise.
Course Author Sagar Udasi
MSc Statistics and Data Science with Computational Finance from The University of Edinburgh.
Contact: sagar.l.udasi@gmail.com
Course Organiser TBD
No. Lecture Title Concepts Covered Lecture Objective
01 Why Your Data Lies More Than You Think Uncertainty, randomness, variability Students understand that data is noisy and decisions require reasoning under uncertainty. Critical for interpreting user data in their AI product.
02 Probability: The Language of Uncertainty Basic probability, events, intuition Gives students a working sense of likelihood, useful for decision-making and model outputs.
03 Averages That Mislead and Averages That Matter Mean, median, variance, standard deviation Helps students summarize data properly and avoid misleading conclusions.
04 The Bell Curve That Runs the World Normal distribution, real-world examples Students recognize common patterns in data and understand what “normal” actually means.
05 When Two Things Move Together (Or Don’t) Correlation, covariance, spurious relationships Prevents students from making false conclusions when analyzing product metrics or model outputs.
06 How Sure Are You, Really? Confidence intervals, estimation Helps students interpret ranges instead of single numbers when evaluating performance.
07 Should You Trust This Result? Hypothesis testing, p-value intuition Gives students a basic filter to judge experiments, A/B tests, and model claims in their projects.
Component Weightage
Short Quiz (Conceptual) 40%
Applied Assignment (Data Interpretation) 40%
In-Class Case Discussion 20%
Type Resource Provider
Lecture Statistics Fundamentals Khan Academy
Lecture Seeing Theory (Interactive) Brown University
Reading Naked Statistics Charles Wheelan
Practical Think Stats (Python-based) Allen B. Downey