Applied Mathematics of Uncertainty
Strengthen your intuition for high-stakes modeling by moving past elementary calculus and probability. Master Bayesian likelihood ratios, Markov spectral dynamics, high-dimensional geometry, and information entropy.
Like Duolingo, but for Applied Mathematics of Uncertainty. Tomo turns the whole topic into a game you play five minutes a day, until it actually sticks.
A short one: 12 levels across 2 sections, about 24 minutes end to end, roughly 5 days at five minutes a day. It moves through Stochastic Dynamics and Belief Updating and High Dimensions and Information Limits.
Free forever · No credit card · iPhone & Android
Key ideas in Applied Mathematics of Uncertainty
- The Bayes factor (likelihood ratio) is P(E|H) / P(E|¬H), which here is 0.90 / 0.05 = 18
- Posterior odds equal prior odds multiplied by the Bayes factor: (1/100) * 18 = 18/100
- Odds form cleanly isolates the test's discriminative strength from the population base rate, avoiding joint probability recalculations
- Multiplying hundreds of small probabilities or likelihood ratios causes floating-point registers to underflow to zero
- Taking the logarithm converts sequential multiplication into stable summation: log(Posterior Odds) = log(Prior Odds) + Σ log(Likelihood Ratio)
- Log-odds (logits) allow linear additive updates as each new piece of conditionally independent evidence arrives
- Optimal Bayesian decisions minimize expected loss E[L(θ, a)] integrated over the full posterior distribution, not the parameter's mode
- Under asymmetric linear loss (pinball/quantile loss), the optimal point decision corresponds to a specific extreme quantile of the posterior, not the mean or mode
- When underestimation penalty dwarfs overestimation penalty (100:1), the optimal action sits far up in the right tail (the ~99th percentile), far above the posterior mean and mode
- A conjugate prior yields a posterior belonging to the same algebraic family as the prior when multiplied by the likelihood
- Conjugate pairs turn Bayesian updates into simple arithmetic additions on hyperparameter counters (e.g., adding successes and failures to Alpha and Beta)
- Closed-form updates eliminate the need for computationally heavy Markov Chain Monte Carlo (MCMC) or numerical quadrature during streaming data ingestion
- Optimal stopping in Bayesian sequential analysis balances the marginal cost of sampling against the expected reduction in loss (Expected Value of Sample Information, EVSI)
- When EVSI < marginal sampling cost, gathering more data destroys net value, dictating an immediate terminal decision
- Information gain exhibits diminishing returns; once residual uncertainty reduction cannot alter the decision or offset sampling costs, collection must stop
- A row-stochastic matrix has a dominant eigenvalue λ₁ = 1 with a trivial right eigenvector of all ones (P 1 = 1)
You've tried the other tabs
Thirty open tabs. Four facts you actually kept.
You watched. You nodded. By Sunday it was gone.
One answer, then back to scrolling.
Eight weeks. You meant to finish. You didn't.
Tomo gives Applied Mathematics of Uncertainty the Duolingo treatment: levels, streaks, and quick quizzes that test what you just learned. That game loop is what the tabs above never had, so it's the one you actually finish.
Here's what playing it feels like
A real question from this course. Take your best guess.
Your lookout sensor spots a defect with a 90% true positive rate and a 5% false positive rate. What is its Bayes factor (likelihood ratio)?
Get it right to open this lesson and 11 more in the app.
Where Applied Mathematics of Uncertainty takes you
- 1
Stochastic Dynamics and Belief Updating
- Calibrating Beliefs with Likelihood Ratios
- Predicting System Convergence with Markov Dynamics
- 2
High Dimensions and Information Limits
- Navigating High-Dimensional Geometric Distortion
- Quantifying Uncertainty with Information Entropy
2 sections · 4 units · 12 levels. Built to play, not to enroll.
Tomo is in ChatGPT
Tomo is also a ChatGPT plugin: the same playable courses, built and played inside your chat.
Add Tomo to ChatGPT- What to ask
- “Make me a Tomo course on black holes.” “Turn this chat into a Tomo course.”
- What you get
- Quick questions with instant feedback, hearts and stars, ready to play in about a minute.
- Where it goes
- No sign-up to start. Connect your Tomo account and your streak carries on in the app.
You pick the voice
Applied Mathematics of Uncertainty is taught in the Pirate style: arr, learning be an adventure!. Want a different feel? In the app you can spin up the same topic in any of Tomo's teaching styles. Same facts, totally different vibe.
More Math on Tomo
Canadian High School Math Assessments
Build the applied problem-solving, modeling, and communication skills tested on provincial high school math and numeracy exams. Learn how to unpack dense contextual prompts, tackle high-yield algebra and geometry scenarios, and write rubric-ready justifications.
Essential Math for Electrical Engineering
Bridge the gap between pure mathematics and hands-on circuit reality before university lectures begin. Build confidence rearranging circuit equations, decoding sinusoidal waveforms, analyzing mechanical forces, and reporting experimental engineering data.
How Mathematics Works
Bridge the gap between mechanical calculations and deep mathematical intuition. Explore functions, calculus, vectors, and formal proofs to see how the mathematical toolkit fits together.
The Math of Change
Bridge the gap between basic algebra and the powerful world of calculus. Learn to predict trends, find optimal solutions, and understand the hidden patterns in how things grow and shrink.
Defining Quantities for Real-World Modeling
Learn how to turn messy, everyday scenarios into clear, computable variables. Discover how to choose meaningful units, build useful rates and ratios, and sanity-check your numbers before running the math.
Applied Mental Arithmetic
Unlock high-speed mental multiplication, non-standard division, and rapid bounding heuristics. Replace pen-and-paper calculation with algebraic levers engineered for everyday quantitative decisions.
Start Applied Mathematics of Uncertainty today.
Download Tomo, search Applied Mathematics of Uncertainty, and play your first lesson in under a minute.