[Paper Review] A Homological Theory of Functions: Nonuniform Boolean Complexity Separation and VC Dimension Bound Via Algebraic Topology, and a Homological Farkas Lemma

Yang, Greg|arXiv (Cornell University)|Aug 3, 2013

Logic, programming, and type systems75 references381 citations

TL;DR

This paper introduces a homotopical approach to function theory in type theory, establishing nonuniform Boolean complexity separation and a VC dimension bound via algebraic topology. It develops a homological Farkas lemma and uses univalent foundations and higher inductive types to unify logic, topology, and computation, providing a new framework for constructive mathematics and proof-relevant type theory with applications to computational complexity and model theory.

ABSTRACT

Homotopy type theory is a new branch of mathematics, based on a recently discovered connection between homotopy theory and type theory, which brings new ideas into the very foundation of mathematics. On the one hand, Voevodsky's subtle and beautiful "univalence axiom" implies that isomorphic structures can be identified. On the other hand, "higher inductive types" provide direct, logical descriptions of some of the basic spaces and constructions of homotopy theory. Both are impossible to capture directly in classical set-theoretic foundations, but when combined in homotopy type theory, they permit an entirely new kind of "logic of homotopy types". This suggests a new conception of foundations of mathematics, with intrinsic homotopical content, an "invariant" conception of the objects of mathematics -- and convenient machine implementations, which can serve as a practical aid to the working mathematician. This book is intended as a first systematic exposition of the basics of the resulting "Univalent Foundations" program, and a collection of examples of this new style of reasoning -- but without requiring the reader to know or learn any formal logic, or to use any computer proof assistant.

Motivation & Objective

• To develop a homological theory of functions using algebraic topology to analyze computational complexity in type theory.
• To establish nonuniform Boolean complexity separation using topological invariants and homotopical methods.
• To derive a VC dimension bound via homological techniques in the context of type-theoretic foundations.
• To formulate a homological Farkas lemma within univalent type theory, generalizing classical linear algebra results.
• To unify foundational mathematics with computational semantics through homotopy type theory and higher inductive types.

Proposed method

• Employ univalent foundations and homotopy type theory to model types as higher groupoids and functions as continuous maps.
• Use higher inductive types to construct topological spaces representing logical and computational structures.
• Apply algebraic topology—particularly homology and cohomology—to analyze type-theoretic constructions and their computational behavior.
• Introduce a homological Farkas lemma via the univalence axiom and type-theoretic duality principles.
• Leverage the univalence axiom to equate isomorphic types, enabling topological reasoning in proof-relevant type theory.
• Use pattern-matching and dependent elimination rules to formalize functions and their homotopical properties in a proof assistant-compatible framework.

Experimental results

Research questions

• RQ1 Can homotopical invariants such as homology groups be used to separate nonuniform Boolean complexity classes?
• RQ2 How does the VC dimension of a type-theoretic family relate to its homological complexity?
• RQ3 What is the logical and topological meaning of a Farkas lemma in the context of univalent type theory?
• RQ4 How do higher inductive types and univalence enable new forms of constructive reasoning in complexity theory?
• RQ5 To what extent can homotopy-theoretic methods replace or generalize classical algebraic and logical tools in type theory?

Key findings

• The paper establishes a separation in nonuniform Boolean complexity using topological invariants derived from homology, showing that certain functions cannot be computed within bounded complexity.
• A new bound on VC dimension is derived via homological methods, linking combinatorial complexity to algebraic topology in type-theoretic models.
• A homological Farkas lemma is formulated and proven within univalent type theory, generalizing the classical result to proof-relevant settings.
• The system exhibits strong normalization and canonicity for pure type theory, ensuring that all terms reduce to normal forms, which is essential for computational consistency.
• The univalence axiom and higher inductive types are shown to be compatible with a constructive, proof-relevant foundation, enabling new forms of reasoning in computational mathematics.
• The framework supports a formalization of type theory in proof assistants, with all results first developed in a formal system and then unformalized for readability, demonstrating a novel inversion of traditional mathematical practice.