A synthetic approach to Markov kernels, conditional independence and theorems on sufficient statistics
arXiv:1908.07021 · doi:10.1016/j.aim.2020.107239
Abstract
We develop Markov categories as a framework for synthetic probability and statistics, following work of Golubtsov as well as Cho and Jacobs. This means that we treat the following concepts in purely abstract categorical terms: conditioning and disintegration; various versions of conditional independence and its standard properties; conditional products; almost surely; sufficient statistics; versions of theorems on sufficient statistics due to Fisher--Neyman, Basu, and Bahadur. Besides the conceptual clarity offered by our categorical setup, its main advantage is that it provides a uniform treatment of various types of probability theory, including discrete probability theory, measure-theoretic probability with general measurable spaces, Gaussian probability, stochastic processes of either of these kinds, and many others.
arXiv comment98 pages. v6: fixed error in Section 7. v7: incorporates referee's comments. v8: minor correction
References in corpus (24)
- Probabilistic theories with purification
- A generalized no-broadcasting theorem
- Disintegration and Bayesian Inversion via String Diagrams
- A Convenient Category for Higher-Order Probability Theory
- Beyond Bell's Theorem II: Scenarios with arbitrary causal structure
- Causal categories: relativistically interacting processes
- Denotational validation of higher-order Bayesian inference
- Picturing classical and quantum Bayesian inference
- Framed bicategories and monoidal fibrations
- From Kleisli Categories to Commutative C*-algebras: Probabilistic Gelfand Duality
- On the computability of conditional probability
- Terminality implies non-signalling
- Statistical topology via Morse theory, persistence and nonparametric estimation
- Bimonoidal Structure of Probability Monads
- Finite matrices are complete for (dagger-)hypergraph categories
- The Beta-Bernoulli process and algebraic effects
- Supplying bells and whistles in symmetric monoidal categories
- Equivalence of relativistic causal structure and process terminality
- Codensity and the Giry monad
- Probabilistic morphisms and Bayesian nonparametrics
- Tensors, monads and actions
- Compositories and Gleaves
- Method of Additional Structures on the Objects of a Monoidal Kleisli Category as a Background for Information Transformers Theory
- Algebras of the extended probabilistic powerdomain monad
Cited by in corpus (69)
- Axioms for retrodiction: achieving time-reversal symmetry with a prior
- Representable Markov Categories and Comparison of Statistical Experiments in Categorical Probability
- On quantum states over time
- A non-commutative Bayes' theorem
- Infinite products and zero-one laws in categorical probability
- Free gs-monoidal categories and free Markov categories
- Bayesian Updates Compose Optically
- Dilations and information flow axioms in categorical probability
- A category-theoretic proof of the ergodic decomposition theorem
- Markov Categories and Entropy
- The information loss of a stochastic map
- Probabilistic morphisms and Bayesian nonparametrics
- Categorial Independence and Lévy Processes
- Parametric models and information geometry on W*-algebras
- Inverses, disintegrations, and Bayesian inversion in quantum Markov categories
- A functorial characterization of von Neumann entropy
- Bayesian inversion and the Tomita-Takesaki modular group
- Category Theory in Machine Learning
- Multinomial and Hypergeometric Distributions in Markov Categories
- Bayesian open games
- Non-commutative disintegrations: existence and uniqueness in finite dimensions
- Learning from What's Right and Learning from What's Wrong
- Coend Optics for Quantum Combs
- Probability monads with submonads of deterministic states - Extended version
- Homotopy Theoretic and Categorical Models of Neural Information Networks
- An Operational Environment for Quantum Self-Testing
- Categorical Stochastic Processes and Likelihood
- Categorical composable cryptography: extended version
- The algebra and machine representation of statistical models
- Conditional Distributions for Quantum Systems
- A Categorical Semantics of Fuzzy Concepts in Conceptual Spaces
- Cyber Kittens, or Some First Steps Towards Categorical Cybernetics
- Open Dynamical Systems as Coalgebras for Polynomial Functors, with Application to Predictive Processing
- A Framework for Universality in Physics, Computer Science, and Beyond
- Probabilistic programming interfaces for random graphs: Markov categories, graphons, and nominal sets
- Functorial Manifold Learning
- Sufficient Statistics and Split Idempotents in Discrete Probability Theory
- Monoidal Width
- Value Iteration is Optic Composition
- From Gs-monoidal to Oplax Cartesian Categories: Constructions and Functorial Completeness
- Hidden Markov Models and the Bayes Filter in Categorical Probability
- A Categorical Construction of the Real Unit Interval
- A category theory framework for Bayesian learning
- Weak cartesian properties of simplicial sets
- Structured and Decorated Cospans from the Viewpoint of Double Category Theory
- On a 2-relative entropy
- Unifilar Machines and the Adjoint Structure of Bayesian Filtering
- Overdrawing Urns using Categories of Signed Probabilities
- Reinforcement Learning in Categorical Cybernetics
- Compositional Semantics for Probabilistic Programs with Exact Conditioning
- A Mathematical Framework for Causally Structured Dilations and its Relation to Quantum Self-Testing
- Lifting couplings in Wasserstein spaces
- Bipartite quantum states admitting a causal explanation
- Compositional Abstraction Error and a Category of Causal Models
- On the categorical foundations of quantum information theory: Categories and the Cramer-Rao inequality
- String Diagrams with Factorized Densities
- Cartesian double theories: A double-categorical framework for categorical doctrines
- A model of stochastic memoization and name generation in probabilistic programming: categorical semantics via monads on presheaf categories
- Functors induced by comma categories
- Monoidal Width: Capturing Rank Width
- Identification of Causal Influences in Quantum Processes
- Drawing with Distance
- De Finetti's construction as a categorical limit
- Biased elementary doctrines and quotient completions
- Quasi-Measurable Spaces: A Convenient Foundation of Probability Theory
- Some Notions of (Open) Dynamical System on Polynomial Interfaces
- Categories of abstract and noncommutative measurable spaces
- Categorical algebra of conditional probability
- The Aldous$\unicode{x2013}$Hoover Theorem in Categorical Probability