# Cosinegate Dynamic Routing

> Achieve efficient neural networks via self-supervised dynamic routing using Cosine Incompatibility Ratio (CIR). Ground gating decisions in geometric novelty rather than learned heuristics, enable per-sample/per-block binary routing via Gumbel-softmax, constrain with progressive FLOPs regularization—maintaining accuracy while reducing computation 28.5% on CIFAR-10.

- Skill: `adu2021/cosinegate-dynamic-routing` (Agent Skill)
- Install (CLI): `npx skillmds@latest add adu2021/cosinegate-dynamic-routing`
- Raw SKILL.md: https://api.skillmd.com/api/skills/adu2021/cosinegate-dynamic-routing/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: adu2021 (https://skillmd.com/u/adu2021)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/adu2021/cosinegate-dynamic-routing

---


## Overview

Dynamic routing based on semantic novelty between identity and residual paths.

## Core Technique

**Cosine Incompatibility Ratio:**

```python
# CIR = 1 - cos(x, F(x))
# Low CIR: identity sufficient (skip block)
# High CIR: block adds novelty (execute)

cir_score = 1.0 - cosine_similarity(input_x, residual_output)
should_skip = cir_score < threshold
```

**Differentiable Gating:**
Gumbel-softmax for training, deterministic for inference.

## Performance

- 91.3% accuracy with 28.5% FLOPs saving
- No auxiliary supervision needed

## References

- Cosine incompatibility ratio for routing
- Geometric grounding of gate decisions
- FLOPs-constrained gating