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.
Fan-in and fan-out target different signal directions
Example
Compare the source's forward and atomless conditional backward recurrence coefficients for fan-in 64 and fan-out 128. The respective coefficients are and . Compute their factors at the two initializer choices and . For a preceding preactivation variance the forward result is a multiple of ; for a downstream second moment the conditional backward result is a multiple of (and of its variance when centered). These coefficient comparisons do not assert a nonzero actual ReLU gradient model satisfying the full backward independence assumptions: at a nondegenerate hidden ReLU layer those assumptions force .
Diagram
The top row represents the forward linear map and ReLU; the lower row shows the reverse-mode maps with . Each box denotes the full vector of the stated dimension, and the arrow labelled denotes the matrix, not a single scalar edge. Thus the forward fan-in is 64 and the backward fan-out is 128. The diagram illustrates these dimensions; the verification below uses the recurrence formulas directly.
Facts & Assumptions
Given: The stated dimensions, source recurrence coefficients, and their conditional interpretations for and .
Forward mode uses weight variance , and backward mode uses . (Rectifier variance-preserving initialization)
The forward preactivation variance factor is . (ReLU forward preactivation-variance recurrence)
Under its stated conditional assumptions, the backward gradient second-moment factor is . (ReLU backward gradient-variance recurrence)
Verification
With fan-in variance , the forward factor is , so is preserved. The backward factor is , so the gradient variance becomes .
With fan-out variance , the backward factor is , so is preserved. The forward factor is , so the next preactivation variance is .
Since the two widths differ, the required variances and differ, and no shared weight variance makes both recurrence coefficients equal to one.
Depends on
Used by
Nothing in the library uses this result yet.
Dependency tree · two levels
14 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
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun, Delving Deep into Rectifiers, §2.2 (standard reference, not scraped)
- Simon J. D. Prince, Understanding Deep Learning, §7.5.1–7.5.3 (standard reference, not scraped)
- Johns Hopkins University, Handout #8: Neural Network Training and Tokenization, §3 (standard reference, not scraped)