Infinite-Dimensional Lie Groups

• Infinite-dimensional Lie groups are smooth manifolds modeled on locally convex spaces, with key examples including diffeomorphism and mapping groups.
• Their regularity properties ensure well-behaved exponential maps and the smooth solvability of time-dependent differential equations.
• Applications span geometric control, gauge theory, and representation theory, effectively extending classical Lie group concepts.

An infinite-dimensional Lie group is a group equipped with a smooth manifold structure modelled on an infinite-dimensional locally convex topological vector space, such that group multiplication and inversion are smooth maps. These groups appear ubiquitously in geometry, representation theory, mathematical physics, and noncommutative geometry, extending the standard finite-dimensional Lie group theory into the infinite-dimensional regime. Key examples include diffeomorphism groups of manifolds, gauge groups in gauge theory, groups of smooth maps between manifolds, direct limits of classical matrix groups, and character groups of topological Hopf algebras.

1. Foundations: Definitions and Structures

Let $G$ be a group endowed with a smooth manifold structure modelled on a locally convex space $E$ (e.g., Banach, Fréchet, or Silva). The maps $m: G \times G \to G$, $m(g, h) := gh$, and $i: G \to G$, $i(g) := g^{-1}$, are required to be smooth in the sense of Bastiani calculus or analogous frameworks for locally convex spaces (Gloeckner et al., 12 Feb 2026). The tangent space at the identity, $T_eG$, becomes the Lie algebra $\mathfrak{g}$, with bracket defined by commutators of left invariant vector fields.

Prominent subclasses include:

• Banach–Lie groups: $E$ is a Banach space, yielding the closest analogues of finite-dimensional Lie theory.
• Fréchet–Lie groups: $E$ is Fréchet, capturing diffeomorphism groups and mapping groups.
• LF- or Silva–Lie groups: Modelled on strict direct limits of Banach spaces, as in groups of real analytic diffeomorphisms ($\Diff^\omega(M)$).

Standard Lie theory concepts such as exponential maps, adjoint representations, and subgroups generalize, but with essential differences. For example, the exponential map need not be globally defined or surjective; submanifolds do not always inherit submanifold structures (Gloeckner et al., 12 Feb 2026).

2. Regularity and Evolution Equations

Regularity concerns the ability to solve and smoothly parametrize time-dependent ODEs in the group. $G$ is called $C^k$-regular if, for any $C^k$ curve $\xi: [0,1] \to \mathfrak{g}$, the initial value problem

$$\gamma'(t) = T_e \ell_{\gamma(t)} \xi(t), \quad \gamma(0) = e$$

has a unique solution $\gamma_\xi$, and the endpoint map $\xi \mapsto \gamma_\xi(1)$ is smooth in the $C^k$ topology (Glockner, 2012). Regularity ensures well-behaved exponential maps and is crucial for functional analytic and representation-theoretic constructions.

Key results:

• All Banach–Lie groups are $L^1$-regular; similar statements hold for many direct limits and mapping groups (Glockner, 2015, Nikitin, 2019).
• Smooth ($C^\infty$) regularity guarantees the efficacy of the Trotter and commutator formulas and robust control-theoretic properties (Glockner, 2015, Glockner et al., 2020).
• The hierarchy

$$C^\infty\text{-regular} \Longleftarrow L^q\text{-regular} \Longleftarrow L^p\text{-regular} \Longleftarrow C^0\text{-regular}$$

allows passage of regularity between function space classes (Glockner, 2015).

In geometric control, regularity ensures the solvability of Carathéodory ODEs with $L^1$ controls on G–manifolds, yielding closure results and bang–bang theorems for reachable sets (Glockner et al., 2020).

3. Key Examples and Constructions

Mapping groups: For a finite-dimensional compact manifold $M$ and finite-dimensional Lie group $K$, $C^\infty(M, K)$ forms a Fréchet–Lie group. For Sobolev ($H^s$) or analytic regularity, one uses Banach or Silva–Lie group models (Gloeckner et al., 12 Feb 2026).

Diffeomorphism groups: $\Diff(M)$ for compact $M$ is a Fréchet–Lie group; for non-compact $M$ one considers $\Diff_c(M)$ (compact support) as an LF–Lie group. The group of real-analytic diffeomorphisms on compact real-analytic $M$ ($\Diff^\omega(M)$) has a Silva–Lie group structure (Glockner, 2016).

Direct limits: For an ascending sequence of finite-dimensional Lie groups, the union $G = \bigcup_n G_n$ inherits a direct-limit Lie group structure if compatible charts exist (Gloeckner et al., 12 Feb 2026).

Character groups of Hopf algebras: For a graded, connected Hopf algebra $\mathscr{H}$ over $K$ and a commutative locally convex algebra $B$, $\mathrm{Hom}_{\rm Alg}(\mathscr{H}, B)$ can be structured as a $K$–analytic regular Lie group (a BCH–Lie group), with the Lie algebra comprising infinitesimal characters (Bogfjellmo et al., 2015).

Groups of germs: For a Banach–Lie group $H$, the group of analytic germs $\mathrm{Germ}(K,H)$ is a regular analytic infinite-dimensional Lie group (Dahmen et al., 2014).

4. Parabolic Subgroups and Representation Theory

The representation theory of infinite-dimensional Lie groups, particularly classical direct-limit groups such as $\SL(\infty,\mathbb{R})$, $\SO(p,\infty)$, and $\Sp(\infty,\mathbb{R})$, fundamentally relies on the structural theory of parabolic subgroups:

• Minimal parabolics are those that contain no proper parabolic subalgebra; for classical groups, they correspond to stabilizers of specific generalized flags in the defining module, with decomposition $P = MAN$ into lim-compact (M), diagonal abelian (A), and unipotent radical (N) (Wolf, 2012).
• Flag-closedness and minimality conditions ensure geometric regularity: $P$ stabilizes a chain of Mackey-closed subspaces; when both conditions hold, maximal lim-compact subgroups act transitively on the flag variety $G/P$, and finite-dimensional theory generalizes (Wolf, 2012).

Induced principal series representations are constructed via amenable induction (using invariant means, not Haar measure): $\Ind_P^G(\tau) = \text{completion of } \bigl\{\psi: G \rightarrow V_\tau \mid \psi(gp) = \tau(p)^{-1}\psi(g),\ \psi\ \text{bounded, RUC}\bigr\}$ with the topology determined by seminorms from invariant means (Wolf, 2012). For minimal flag-closed $P$, there exists a natural decomposition of these representations on restriction to maximal lim-compact $K$, given by $(\Ind_P^G \tau)|_K \simeq \Ind_M^K(\tau|_M)$.

5. Topological and Completeness Properties

A critical topological property for robust Lie theory is Weil completeness (completeness in the left uniformity). Glöckner showed that every infinite-dimensional Lie group modelled on a Silva space (strict direct limit of Banach spaces with compact inclusions) is locally $k_\omega$ and thus Weil complete (Glockner, 2016). This ensures the convergence of left-Cauchy nets and the feasibility of constructing flows and integrating vector fields.

Silva–modelled Lie groups include:

• $\Diff^\omega(M)$ for compact real-analytic $M$
• Mapping groups $C^\omega(M,G)$
• Direct limits of finite-dimensional Lie groups

Completeness provides a foundation for integrating Lie algebra elements, constructing representations, and ensuring no “escape to infinity” for solution flows.

6. Smooth Vectors, Smoothing Operators, and Representations

Given a unitary representation $(\pi, \mathcal{H})$ of an infinite-dimensional Lie group $G$, analysis often centers on the space of smooth vectors $\mathcal{H}^{\infty}$. A vector $v$ is smooth if the orbit map $G \to \mathcal{H}$, $g \mapsto \pi(g)v$, is smooth in the sense of locally convex analysis.

Key results (Neeb, 2010, Neeb et al., 2015):

• For Banach–Lie groups, $\mathcal{H}^{\infty}$ is a Fréchet space with a natural projective limit topology.
• Smoothness can be characterized by the smoothness of the positive-definite function $\varphi_v(g) = \langle \pi(g)v, v \rangle$ near the identity ("one-point test").
• Smoothing operators ($A$ with $A(\mathcal{H}) \subset \mathcal{H}^{\infty}$) allow the construction of host $C^*$-algebras, supporting classification of semibounded representations.
• For semibounded representations, $\mathcal{H}^{\infty}$ coincides with the space of smooth vectors for a suitable one-parameter subgroup.

Pathologies can arise: there exist representations where the space of $C^{k+1}$-vectors is zero but $C^k$-vectors are dense, reflecting the subtleties peculiar to infinite-dimensional groups (Neeb, 2010).

7. Cohomological and Geometric Invariants

Characteristic class theory extends partially to infinite-dimensional Lie groups via Chern–Weil theory, leveraging invariant polynomials constructed from traces on certain subalgebras (e.g., gauge, pseudodifferential, or symbol traces) (Rosenberg, 2013):

• On gauge groups, fiberwise traces and integration over the base manifold produce nontrivial classes.
• On pseudodifferential groups, the Wodzicki residue and leading-symbol trace provide candidate invariants, though their universality and nontriviality remain an open area of investigation.
• For diffeomorphism groups, no nontrivial Ad-invariant polynomials are currently known, highlighting significant gaps vs. finite-dimensional theory.

Applications include detection of nontrivial cohomology in mapping spaces, construction of secondary Chern–Simons classes, and partial Chern–Weil–type results for infinite-rank index theorems.

---

These aspects combine to form the central architecture of current research into infinite-dimensional Lie groups: robust geometric foundations, analytic regularity, topological completeness, representation theory extensions, and generalizations of classical invariants, with many foundational open problems and applications across mathematics and physics. For a systematic treatment and further classes (e.g., half-Lie groups, pro-Lie groups), see the references indicated in each section.