feat: GF-T 2-layer ReLU network (solves XOR, proven on AX7203) - #1825
Merged
Conversation
gft_xornet.t27: 2-layer ReLU forward pass h=relu(W*x+c), y=v.h+b. A single linear model cannot separate XOR; the hidden ReLU layer makes it nonlinearly separable. Reuses smul/sadd/relu. Proven on a live AX7203 (uart_xornet.v, analytic XOR weights baked): XOR truth table 4/4 correct AND the correct nonlinear tent surface (peak 1.0 at x0+x1=1, 0 at 0/2) -- multi-layer nonlinear inference on silicon. fasm 9.4M, under the measured correctness ceiling. In-spec tests (4 corners) PASS. docs/NOW.md updated. Refs #1764 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Contributor
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
New spec gft_xornet.t27 — 2-layer ReLU forward pass h=relu(W*x+c), y=v.h+b. A single linear model cannot separate XOR; the hidden ReLU layer makes it nonlinearly separable. In-spec tests (4 XOR corners) PASS. Proven on a live AX7203 (uart_xornet.v, analytic XOR weights baked): XOR truth table 4/4 + the correct nonlinear tent surface — multi-layer nonlinear inference on silicon. fasm 9.4M, under the correctness ceiling. docs/NOW.md updated. Refs #1764