Giry monad¶

The probability monad on measurable spaces that sends each space to its measurable space of probability measures.

Version: v2 · 2026-09-19 · History
Domain-specific #: 4733
Domain group: Formal Sciences
Origin domain: Mathematics
Subdomain: Categorical Probability → Mathematics

Core Idea¶

The sigma-algebra on measures is generated by evaluation maps, variants use subprobability or restricted measurable categories and composition must preserve measurability; it is not merely a set-valued distribution constructor. The unit maps a point to its Dirac measure, while multiplication integrates a probability measure over probability measures into one barycentric measure; Kleisli arrows are Markov kernels. The abstraction is therefore identified by a declared carrier, a transformation or constraint over that carrier, and an invariant that tells an analyst whether the named structure is genuinely present.

How would you explain it like I'm…

No faithful explanation at this level. At this level it collapses into 'a bag of possibilities with chances', a plain distribution-maker. That loses the measurability structure and the lawful composition of random steps that define the Giry monad.

The Spinner-of-Spinners Recipe

Imagine a spinner that tells you the chances of landing on different things. The Giry monad is a mathematical recipe that, for any collection of things, builds the collection of all possible spinners over it. A single sure thing becomes a spinner that always lands on that thing. A 'spinner of spinners' (first spin to pick a spinner, then spin that one) can be squashed into one ordinary spinner by averaging. The recipe also comes with careful rules about which questions you are allowed to ask about chances, and those rules are part of what it is.

The Probability-Measure Monad

The Giry monad lives in the category of measurable spaces, which are sets equipped with a sigma-algebra of events you can assign probabilities to. It sends each space X to the space of probability measures on X, and that new space gets its own sigma-algebra generated by the maps 'evaluate this measure on event A'. Measurable functions are pushed forward to act on measures. The unit turns a point x into the Dirac measure concentrated at x, and the multiplication takes a random probability measure and integrates it into one averaged (barycentric) measure. Its 'Kleisli arrows', maps from X to measures on Y, are exactly Markov kernels, i.e. random transitions, and composing them is like chaining random steps. Variants use subprobability measures or restricted categories of spaces.

The Giry monad is a monad on the category Meas of measurable spaces. Its object map sends X to P(X), the set of probability measures on X, with the sigma-algebra generated by the evaluation maps μ ↦ μ(A) for measurable A; on morphisms it acts by pushforward, making P a functor. The unit η_X: X → P(X) sends x to the Dirac measure δ_x, and the multiplication μ_X: P(P(X)) → P(X) integrates a measure over measures into its barycenter, (μ_X M)(A) = ∫ ν(A) dM(ν). These satisfy the monad unit and associativity laws, and every construction must preserve measurability. Kleisli morphisms X → P(Y) are Markov kernels, and Kleisli composition is the Chapman–Kolmogorov composition of kernels. Further structure, such as strength and commutativity, needs to be stated with its qualifications, and there are subprobability and restricted-category variants. It is therefore a precise categorical object, not just any rule that assigns a set of distributions.

Scope of Application¶

Giry monad belongs to categorical probability and is useful where the analyst can specify the typed categorical probability carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets, then evaluate the category of measurable spaces, object map X to probability measures on X, evaluation-generated sigma-algebra, functorial pushforward, Dirac unit, multiplication by integration, monad unit and associativity laws, Kleisli morphisms and Markov-kernel composition, strength or commutativity qualifications and probability-versus-subprobability variants are explicit.

Clarity¶

The abstraction clarifies a crowded vocabulary by making the category of measurable spaces, object map X to probability measures on X, evaluation-generated sigma-algebra, functorial pushforward, Dirac unit, multiplication by integration, monad unit and associativity laws, Kleisli morphisms and Markov-kernel composition, strength or commutativity qualifications and probability-versus-subprobability variants are explicit the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test.

Manages Complexity¶

Without the abstraction, an analyst must reason directly over many local details: the carrier roles, admissibility assumptions, competing conventions, derived invariants, boundary cases, and proof or validation obligations specific to Giry monad. Giry monad compresses them into the roles in the structural signature. That compression permits comparison across instances without erasing the variables that determine validity. It also exposes which details may be varied safely and which are constitutive.

Abstract Reasoning¶

Identify the carrier. State what the elements, states, objects, or observations are: the typed categorical probability carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets. Reject examples whose alleged carrier belongs to a different problem.
Lock the constitutive rule. Express the category of measurable spaces, object map X to probability measures on X, evaluation-generated sigma-algebra, functorial pushforward, Dirac unit, multiplication by integration, monad unit and associativity laws, Kleisli morphisms and Markov-kernel composition, strength or commutativity qualifications and probability-versus-subprobability variants are explicit independently of one notation or implementation.

Knowledge Transfer¶

Knowledge transfers strongly among subfields of categorical probability because they reuse the typed categorical probability carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets, The unit maps a point to its Dirac measure, while multiplication integrates a probability measure over probability measures into one barycentric measure; Kleisli arrows are Markov kernels., and type the carrier, state every parameter and convention in the definition, test that the category of measurable spaces, object map X to probability measures on X, evaluation-generated sigma-algebra, functorial pushforward, Dirac unit, multiplication by integration, monad unit and associativity laws, Kleisli morphisms and Markov-kernel composition, strength or commutativity qualifications and probability-versus-subprobability variants are explicit, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases.

Relationships to Other Abstractions¶

Current abstraction Giry monad Domain-specific

Parents (1) — outgoing typed relations

Giry monad is a kind of Recursion Prime

It remains its own entry because its identity is fixed by the category of measurable spaces, object map X to probability measures on X, evaluation-generated sigma-algebra, functorial pushforward, Dirac unit, multiplication by integration, monad unit and associativity laws, Kleisli morphisms and Markov-kernel composition, strength or commutativity qualifications and probability-versus-subprobability variants.

Hierarchy path (1) — routes to 1 parentless root

Giry monad → is a kind of Recursion

Neighborhood in Abstraction Space¶

Giry monad sits in a crowded region of the domain-specific corpus (29^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Categorical Constructions & Dualities (33 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-25