How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Pointwise nonlinearities, neighborhood pooling, subgroup sampling, and coset pooling
Definition
Let be a group and let be a feature map with channels. A pointwise nonlinearity determined by is , where . For a nonempty finite set , non-strided neighborhood max pooling is , where ; the output is still indexed by . For a subgroup , subgroup sampling is the restriction , indexed by .
The left-coset set is , where . Its left -action is . For finite , left-coset max pooling is the map given by . All maxima are taken separately in each channel. When is finite, has one element and is a group-invariant summary.
Definition references
left actions on group-indexed feature maps
left actions of groups on sets
subgroup identity, closure, and inverse conditions
left cosets of a subgroup
functions and declared domains/codomains
Cohen and Welling, “Group Equivariant Convolutional Networks,” §§6.2–6.3
Goodfellow, Bengio, and Courville, Deep Learning, Chapter 9 §9.3
Facts & Assumptions
Given: A group , a finite subgroup , a nonempty finite , and a feature map . Verify that the coset action and pooling formulas are well-defined.
Subgroup: A subgroup contains the identity and is closed under multiplication and inverses.
Left and right cosets and of a subgroup: The left coset represented by is .
Left group actions, transitive actions, and faithful actions: A left action satisfies and for each acted-on element .
Proof
If , then , so [F2] gives for some . By [F1], ; conversely, every equals with , so . Hence and , proving that is independent of the representative. It satisfies the identity and composition laws in [F3] because and . Thus it is a left action on .
If with , then : closure gives , and for , by [F1]. Hence , so the componentwise maximum defining is independent of the representative.
By [F1], contains the identity, so and the given set are nonempty and finite. Each displayed maximum therefore exists in ; the pointwise operation and subgroup restriction are also functions on their stated domains. Thus all four operators in the Definition are well-defined.
Depends on
- The left action on feature maps induced by an action on sites
- Left group actions, transitive actions, and faithful actions
- Group and abelian group
- Subgroup
- Left and right cosets $gH$ and $Hg$ of a subgroup
- A function is a relation $f$ with $(a,b) \in f$ and $(a,c) \in f$ implying $b = c$; $f : A \to B$, the value $f(a)$, domain and codomain
Used by
Dependency tree · two levels
16 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.
Sources
- Taco S. Cohen and Max Welling, Group Equivariant Convolutional Networks (standard reference, not scraped)
- Ian Goodfellow, Yoshua Bengio, and Aaron Courville, Deep Learning (standard reference, not scraped)