ruvector/examples/spiking-network/src/lib.rs
rUv 6c00b84e1d
feat(micro-hnsw-wasm): Add Neuromorphic HNSW v2.3 with SNN Integration (#40)
* docs: Add comprehensive GNN v2 implementation plans

Add 22 detailed planning documents for 19 advanced GNN features:

Tier 1 (Immediate - 3-6 months):
- GNN-Guided HNSW Routing (+25% QPS)
- Incremental Graph Learning/ATLAS (10-100x faster updates)
- Neuro-Symbolic Query Execution (hybrid neural + logical)

Tier 2 (Medium-Term - 6-12 months):
- Hyperbolic Embeddings (Poincaré ball model)
- Degree-Aware Adaptive Precision (2-4x memory reduction)
- Continuous-Time Dynamic GNN (concept drift detection)

Tier 3 (Research - 12+ months):
- Graph Condensation (10-100x smaller graphs)
- Native Sparse Attention (8-15x GPU speedup)
- Quantum-Inspired Attention (long-range dependencies)

Novel Innovations (10 experimental features):
- Gravitational Embedding Fields, Causal Attention Networks
- Topology-Aware Gradient Routing, Embedding Crystallization
- Semantic Holography, Entangled Subspace Attention
- Predictive Prefetch Attention, Morphological Attention
- Adversarial Robustness Layer, Consensus Attention

Includes comprehensive regression prevention strategy with:
- Feature flag system for safe rollout
- Performance baseline (186 tests + 6 search_v2 tests)
- Automated rollback mechanisms

Related to #38

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

* feat(micro-hnsw-wasm): Add neuromorphic HNSW v2.3 with SNN integration

## New Crate: micro-hnsw-wasm v2.3.0
- Published to crates.io: https://crates.io/crates/micro-hnsw-wasm
- 11.8KB WASM binary with 58 exported functions
- Neuromorphic vector search combining HNSW + Spiking Neural Networks

### Core Features
- HNSW graph-based approximate nearest neighbor search
- Multi-distance metrics: L2, Cosine, Dot product
- GNN extensions: typed nodes, edge weights, neighbor aggregation
- Multi-core sharding: 256 cores × 32 vectors = 8K total

### Spiking Neural Network (SNN)
- LIF (Leaky Integrate-and-Fire) neurons with membrane dynamics
- STDP (Spike-Timing Dependent Plasticity) learning
- Spike propagation through graph topology
- HNSW→SNN bridge for similarity-driven neural activation

### Novel Neuromorphic Features (v2.3)
- Spike-Timing Vector Encoding (rate-to-time conversion)
- Homeostatic Plasticity (self-stabilizing thresholds)
- Oscillatory Resonance (40Hz gamma synchronization)
- Winner-Take-All Circuits (competitive selection)
- Dendritic Computation (nonlinear branch integration)
- Temporal Pattern Recognition (spike history matching)
- Combined Neuromorphic Search pipeline

### Performance Optimizations
- 5.5x faster SNN tick (2,726ns → 499ns)
- 18% faster STDP learning
- Pre-computed reciprocal constants
- Division elimination in hot paths

### Documentation & Organization
- Reorganized docs into subdirectories (gnn/, implementation/, publishing/, status/)
- Added comprehensive README with badges, SEO, citations
- Added benchmark.js and test_wasm.js test suites
- Added DEEP_REVIEW.md with performance analysis
- Added Verilog RTL for ASIC synthesis

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-12-01 22:30:15 -05:00

71 lines
2.1 KiB
Rust

//! # Spiking Neural Network Library
//!
//! Event-driven spiking neural network implementation optimized for ASIC deployment.
//!
//! ## Philosophy
//!
//! Spiking neural networks do not compute in the traditional sense. They fire only when
//! something meaningful happens. Everything is event-driven. This single shift changes
//! the entire energy and timing model of your ASIC.
//!
//! A conventional network evaluates every neuron every cycle. It burns power on
//! multiplications even when nothing is changing. A spiking model skips all of that.
//! Neurons stay silent until a threshold is crossed. You only compute on change.
//!
//! ## Architecture Benefits
//!
//! - **Sparse computation**: Only active neurons consume resources
//! - **Event-driven**: No wasted cycles on unchanged state
//! - **Local connectivity**: Minimizes routing complexity
//! - **Tiny events**: Each spike is just a few bits
//! - **Microsecond latency**: Local lookups instead of matrix multiplies
//!
//! ## Usage
//!
//! ```rust,ignore
//! use spiking_network::{
//! neuron::{LIFNeuron, NeuronParams},
//! network::SpikingNetwork,
//! encoding::SpikeEncoder,
//! };
//!
//! // Create a network with 1000 neurons
//! let mut network = SpikingNetwork::new(1000);
//!
//! // Encode input as sparse spikes
//! let spikes = SpikeEncoder::rate_encode(&input_data, 0.1);
//!
//! // Process - only fires on meaningful events
//! let output = network.process(&spikes);
//! ```
#![warn(missing_docs)]
#![deny(unsafe_op_in_unsafe_fn)]
pub mod encoding;
pub mod error;
pub mod learning;
pub mod network;
pub mod neuron;
pub mod router;
// Re-exports for convenience
pub use encoding::{SpikeEncoder, SpikeEvent, SpikeTrain};
pub use error::{Result, SpikingError};
pub use learning::{STDPConfig, STDPLearning};
pub use network::{NetworkConfig, NetworkStats, SpikingNetwork};
pub use neuron::{IzhikevichNeuron, LIFNeuron, NeuronParams, SpikingNeuron};
pub use router::{AsicRouter, RouterConfig, SpikePacket};
/// Library version
pub const VERSION: &str = env!("CARGO_PKG_VERSION");
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_version() {
assert!(!VERSION.is_empty());
}
}