Crate ruvector_domain_expansion 

Domain Expansion Engine

Cross-domain transfer learning for general problem-solving capability.

Core Insight

True IQ growth appears when a kernel trained on Domain 1 improves Domain 2 faster than Domain 2 alone. That is generalization.

Two-Layer Architecture

Policy learning layer: Meta Thompson Sampling with Beta priors across context buckets. Chooses strategies via uncertainty-aware selection. Transfer happens through compact priors — not raw trajectories.

Operator layer: Deterministic domain kernels (Rust synthesis, planning, tool orchestration) that generate tasks, evaluate solutions, and produce embeddings into a shared representation space.

Domains

• Rust Program Synthesis: Generate Rust functions from specifications
• Structured Planning: Multi-step plans with dependencies and resources
• Tool Orchestration: Coordinate multiple tools/agents for complex goals

Transfer Protocol

1. Train on Domain 1, extract TransferPrior (posterior summaries)
2. Initialize Domain 2 with dampened priors from Domain 1
3. Measure acceleration: cycles to convergence with/without transfer
4. A delta is promotable only if it improves target without regressing source

Population-Based Policy Search

Run a population of PolicyKernel variants in parallel. Each variant tunes knobs (skip mode, prepass, speculation thresholds). Keep top performers on holdouts, mutate, repeat.

Acceptance Test

Domain 2 must converge faster than Domain 1 to target accuracy, cost, robustness, and zero policy violations.

Re-exports

pub use cost_curve::AccelerationEntry;

pub use cost_curve::AccelerationScoreboard;

pub use cost_curve::ConvergenceThresholds;

pub use cost_curve::CostCurve;

pub use cost_curve::CostCurvePoint;

pub use cost_curve::ScoreboardSummary;

pub use domain::Domain;

pub use domain::DomainEmbedding;

pub use domain::DomainId;

pub use domain::Evaluation;

pub use domain::Solution;

pub use domain::Task;

pub use planning::PlanningDomain;

pub use policy_kernel::PolicyKernel;

pub use policy_kernel::PolicyKnobs;

pub use policy_kernel::PopulationSearch;

pub use policy_kernel::PopulationStats;

pub use rust_synthesis::RustSynthesisDomain;

pub use tool_orchestration::ToolOrchestrationDomain;

pub use meta_learning::CuriosityBonus;

pub use meta_learning::DecayingBeta;

pub use meta_learning::MetaLearningEngine;

pub use meta_learning::MetaLearningHealth;

pub use meta_learning::ParetoFront;

pub use meta_learning::ParetoPoint;

pub use meta_learning::PlateauAction;

pub use meta_learning::PlateauDetector;

pub use meta_learning::RegretSummary;

pub use meta_learning::RegretTracker;

pub use transfer::ArmId;

pub use transfer::BetaParams;

pub use transfer::ContextBucket;

pub use transfer::DualPathResult;

pub use transfer::MetaThompsonEngine;

pub use transfer::TransferPrior;

pub use transfer::TransferVerification;

Modules

cost_curve

Cost Curve Compression Tracker and Acceleration Scoreboard

domain

Core domain trait and types for cross-domain transfer learning.

meta_learning

Meta-Learning Improvements for AGI Learning Architecture

planning

Structured Planning Tasks Domain

policy_kernel

PolicyKernel: Population-Based Policy Search

rust_synthesis

Rust Program Synthesis Domain

tool_orchestration

Tool Orchestration Problems Domain

transfer

Cross-Domain Transfer Engine with Meta Thompson Sampling

Structs

DomainExpansionEngine

The domain expansion orchestrator.