Explore the topics

Topics: from statistics to language models

6 subject areas, 23 guides and 20 interactive tools. Explore a specific question or follow a learning path, from basic ideas to decisions in practice.

Find your starting point

Four paths through the methods

Start with a question. Each guide stands on its own; the order is a suggestion, with no account or required steps.

5 articles

Bayes & statistics

Learn from data without hiding uncertainty: prior, likelihood and posterior, and what they mean for experiments.

Interactive: Bayesian update visualiser, Bayesian A/B test calculator, Bootstrap: estimating uncertainty by resampling, Calibration: can predicted probabilities be trusted?, Causal inference: separating effects from associations

Bayes & statistics

Bayesian statistics explained

How reliable is a success rate with little data? Understand prior, likelihood and posterior through a business example and try the Bayesian calculator.

10 min read· with an interactive explainer
Bayes & statistics

Bayesian A/B testing

How likely is B to be better, and what would a mistake cost? Beta posteriors, expected loss, uplift intervals and honest stopping rules.

10 min read· with an interactive explainer
Bayes & statistics

Bootstrap: estimating uncertainty by resampling

Bootstrap explained: compare mean and median, resample observations with replacement and explore a confidence interval in an interactive local example.

4 min read· with an interactive explainer
Bayes & statistics

Calibration: can predicted probabilities be trusted?

Understand probability calibration through a reliability diagram, Brier score and temperature correction using interactive, fully synthetic predictions.

4 min read· with an interactive explainer

6 articles

Decisions & optimisation

Expected value, utility, decision trees and mathematical optimisation: from weighing alternatives to the best feasible action.

Interactive: Value of information: when is a test worthwhile?, Bayesian optimisation: which experiment comes next?, Linear optimisation, Robust decisions

Decisions & optimisation

Decision theory and decision trees

From payoff matrices through expected value and expected utility to decision trees: comparing options under uncertainty—and valuing additional information.

11 min read
Decisions & optimisation

Optimization: linear programs and scheduling

Decision variables, objectives and constraints: how linear programs work, what shadow prices reveal, why rounding fails and how scheduling rules affect results.

12 min read
Decisions & optimisation

Bayesian optimisation: which experiment comes next?

Bayesian optimisation explained: explore a real Gaussian process, model uncertainty and expected improvement while searching for a good parameter setting.

4 min read· with an interactive explainer
Decisions & optimisation

Linear optimisation: allocate limited capacity

How should limited labour and material be shared between two products? A small production model shows how constraints shape feasible plans and why whole units can change the answer.

4 min read· with an interactive explainer
Decisions & optimisation

Value of information: when is a test worthwhile?

Understand EVPI and EVSI through a market-test decision tree. Compare uncertainty, test accuracy and cost in an interactive example with explicit assumptions.

3 min read· with an interactive explainer
Decisions & optimisation

Robust decisions: expected value or downside protection?

How much should you order when demand is uncertain? The same numbers produce different quantities under different decision rules. This calculator makes explicit which uncertainty and which kind of loss each rule addresses.

4 min read· with an interactive explainer

2 articles

Simulation & uncertainty

When formulas are not enough: use randomness deliberately, explore scenarios and make risks visible.

Interactive: Monte Carlo demo, Queue simulation

Simulation & uncertainty

Monte Carlo simulation explained

How likely is a project to meet its deadline? Simulate durations, understand P80 and identify key risks, with a worked example and free calculator.

11 min read· with an interactive explainer
Simulation & uncertainty

Queue simulation: utilisation and waiting time

A team can have enough average capacity and still create long queues. This simulation explores how utilisation, variable service times and a second server interact.

4 min read· with an interactive explainer

6 articles

Machine learning & feedback

From regression and clustering to bandits and reinforcement learning: recognise patterns, test predictions and learn from consequences.

Interactive: Bandit simulation, Gridworld (value iteration & Q-learning), Contextual bandits, Reward design, Regression and overfitting, k-means clustering

Machine learning & feedback

Multi-armed bandits & Thompson sampling

Use the best option or try something new? Regret, ε-greedy, UCB1 and Thompson sampling, with a worked step and an interactive simulation.

9 min read· with an interactive explainer
Machine learning & feedback

Reinforcement Learning explained

How an agent learns by acting and receiving feedback: rewards, Markov decision processes, the Bellman equation, value iteration and Q-learning—with a worked example.

10 min read· with an interactive explainer
Machine learning & feedback

Contextual bandits

Why the same action can perform differently in different situations.

4 min read· with an interactive explainer
Machine learning & feedback

Reward design

An optimal route can produce a poor outcome when its reward ignores important consequences.

4 min read· with an interactive explainer
Machine learning & feedback

Regression and overfitting: training is not testing

A curve can fit familiar observations extremely well and predict new ones poorly. Compare polynomial degrees on the same training and holdout data, and see when additional flexibility becomes harmful.

4 min read· with an interactive explainer
Machine learning & feedback

k-means clustering: why feature scaling matters

Which customers are similar? The answer also depends on how similarity is measured. Group the same synthetic customers by order count and monthly revenue, and explore the effect of scaling.

4 min read· with an interactive explainer

2 articles

Language models & knowledge

See how attention combines information and retrieval supplies sources. Learn where a language model needs checks and explicit rules.

Interactive: Transformer attention, RAG and sources

Language models & knowledge

Transformer attention

How similarity scores become weights and weighted values become a new representation.

4 min read· with an interactive explainer
Language models & knowledge

RAG and sources

Finding a relevant source is a separate calculation, and does not yet produce a verified answer.

4 min read· with an interactive explainer

2 articles

Decision intelligence

How statistics, simulation, optimisation and learning combine into an understandable decision system.

Interactive: Rules and optimisation

Decision intelligence

What is Decision Intelligence?

How data becomes decisions we can evaluate: a definition, the distinction from BI and Predictive Analytics, the methods involved and a fully worked example.

13 min read
Decision intelligence

Rules and optimisation

A plausible schedule must satisfy its rules before its value can be optimised.

4 min read· with an interactive explainer

From NyxAI

smartlytics is the learning hub of NyxAI GmbH.

Explore the methods we work with, explained openly, with sources and calculations you can reproduce.

Decision Core

NyxAI is developing a Decision Core for Decision Intelligence that combines optimisation, simulation, statistics and reinforcement learning: compare options, make uncertainty visible and learn from outcomes, with people in control.

AI workflows at NyxAI

DeepSquare

Explainable chess AI with human playing styles: NyxAI’s testing ground for understandable decisions under time pressure.