Skip to main content

📝 Greek Letters

Greek symbols are conventions, not fixed definitions. Their meaning depends on the surrounding paper, code, or model, and uppercase and lowercase forms can mean different things.

The IPA below reflects common English mathematical pronunciation; regional variants are noted where especially common.

Greek letterEnglish namePronunciation (IPA)Common ML meaningsExample / notes
Α αalpha/ˈæl.fə/Learning rate, mixing coefficient, distribution shape parameterα = 10⁻³ may be a learning rate
Β βbetaUS /ˈbeɪ.tə/, UK /ˈbiː.tə/Regression coefficient, Beta-distribution parameter, momentum coefficientβ is often a learned linear-model coefficient
Γ γgamma/ˈɡæm.ə/RL discount factor, kernel scale, normalization scaleγ = 0.99 is common in reinforcement learning
Δ δdelta/ˈdel.tə/Change, difference, residual, Kronecker deltaΔw can denote a parameter update
Ε ε, ϵepsilon/ˈep.sɪ.lɒn/Numerical-stability constant, small perturbation, RL exploration rateε = 10⁻⁸ can prevent division by zero
Ζ ζzeta/ˈzeɪ.tə/Rare generic parameterIt has no widely fixed ML meaning
Η ηeta/ˈeɪ.tə/Learning rateηₜ often denotes the learning rate at step t
Θ θ, ϑtheta/ˈθeɪ.tə/Model parametersθ commonly represents all learnable parameters
Ι ιiota/aɪˈoʊ.tə/Rare index or indicatorIt has no widely fixed ML meaning
Κ κkappa/ˈkæp.ə/Cohen's kappa metric, concentration parameterκ is the concentration parameter of a von Mises–Fisher distribution
Λ λlambda/ˈlæm.də/Regularization coefficient, eigenvalue, rate parameterL + λ‖w‖² adds an L2 penalty
Μ μmuUK /mjuː/, US /muː/Mean, momentum coefficientx ∼ 𝒩(μ, σ²) uses μ as the mean
Ν νnu/nuː/Degrees of freedom, ν-SVM parameterIts meaning is usually defined locally
Ξ ξxi/ksaɪ/Random variable, noise sample, generic vectorξ ∼ 𝒩(0, I) can denote sampled noise
Ο οomicron/ˈɑː.mɪ.krɑːn/Rare generic parameterOften avoided because it resembles Latin o and zero
Π π, ϖpi/paɪ/Class prior, probability distribution, RL policy; product (uppercase)πθ(a ∣ s) is a policy's action probability
Ρ ρ, ϱrho/roʊ/Correlation coefficient, density or ratioρ(X, Y) can denote correlation
Σ σ, ςsigma/ˈsɪɡ.mə/Standard deviation, sigmoid function; summation or covariance (uppercase)p = σ(z) can be a probability, but σ alone does not mean probability
Τ τtau/taʊ/Temperature, time constant, time stepsoftmax(z/τ) uses temperature τ
Υ υupsilon/ˈʌp.sɪ.lɒn/Rare generic variableIt has no fixed common ML meaning
Φ φ, ϕphi/faɪ/Feature map, probability density or CDF, generic functionφ(x) may be a feature map; Φ(x) often denotes a normal CDF
Χ χchi/kaɪ/Chi-square statistic or distributionχ² is used in some statistical tests
Ψ ψpsi/psaɪ/Generic function; occasionally the digamma functionIts meaning is usually defined locally
Ω ωomega/oʊˈmeɡ.ə/Generic variable; sample space or parameter set (uppercase)It has no universal ML meaning