d14150da75
Implement intelligent agent learning from Knowledge Graph execution history with per-task-type expertise tracking, recency bias, and learning curves. ## Phase 5.3 Implementation ### Learning Infrastructure (✅ Complete) - LearningProfileService with per-task-type expertise metrics - TaskTypeExpertise model tracking success_rate, confidence, learning curves - Recency bias weighting: recent 7 days weighted 3x higher (exponential decay) - Confidence scoring prevents overfitting: min(1.0, executions / 20) - Learning curves computed from daily execution windows ### Agent Scoring Service (✅ Complete) - Unified AgentScore combining SwarmCoordinator + learning profiles - Scoring formula: 0.3*base + 0.5*expertise + 0.2*confidence - Rank agents by combined score for intelligent assignment - Support for recency-biased scoring (recent_success_rate) - Methods: rank_agents, select_best, rank_agents_with_recency ### KG Integration (✅ Complete) - KGPersistence::get_executions_for_task_type() - query by agent + task type - KGPersistence::get_agent_executions() - all executions for agent - Coordinator::load_learning_profile_from_kg() - core KG→Learning integration - Coordinator::load_all_learning_profiles() - batch load for multiple agents - Convert PersistedExecution → ExecutionData for learning calculations ### Agent Assignment Integration (✅ Complete) - AgentCoordinator uses learning profiles for task assignment - extract_task_type() infers task type from title/description - assign_task() scores candidates using AgentScoringService - Fallback to load-based selection if no learning data available - Learning profiles stored in coordinator.learning_profiles RwLock ### Profile Adapter Enhancements (✅ Complete) - create_learning_profile() - initialize empty profiles - add_task_type_expertise() - set task-type expertise - update_profile_with_learning() - update swarm profiles from learning ## Files Modified ### vapora-knowledge-graph/src/persistence.rs (+30 lines) - get_executions_for_task_type(agent_id, task_type, limit) - get_agent_executions(agent_id, limit) ### vapora-agents/src/coordinator.rs (+100 lines) - load_learning_profile_from_kg() - core KG integration method - load_all_learning_profiles() - batch loading for agents - assign_task() already uses learning-based scoring via AgentScoringService ### Existing Complete Implementation - vapora-knowledge-graph/src/learning.rs - calculation functions - vapora-agents/src/learning_profile.rs - data structures and expertise - vapora-agents/src/scoring.rs - unified scoring service - vapora-agents/src/profile_adapter.rs - adapter methods ## Tests Passing - learning_profile: 7 tests ✅ - scoring: 5 tests ✅ - profile_adapter: 6 tests ✅ - coordinator: learning-specific tests ✅ ## Data Flow 1. Task arrives → AgentCoordinator::assign_task() 2. Extract task_type from description 3. Query KG for task-type executions (load_learning_profile_from_kg) 4. Calculate expertise with recency bias 5. Score candidates (SwarmCoordinator + learning) 6. Assign to top-scored agent 7. Execution result → KG → Update learning profiles ## Key Design Decisions ✅ Recency bias: 7-day half-life with 3x weight for recent performance ✅ Confidence scoring: min(1.0, total_executions / 20) prevents overfitting ✅ Hierarchical scoring: 30% base load, 50% expertise, 20% confidence ✅ KG query limit: 100 recent executions per task-type for performance ✅ Async loading: load_learning_profile_from_kg supports concurrent loads ## Next: Phase 5.4 - Cost Optimization Ready to implement budget enforcement and cost-aware provider selection.
69 lines
3.7 KiB
Plaintext
69 lines
3.7 KiB
Plaintext
-- Migration 005: Knowledge Graph Persistence
|
|
-- Stores execution history and analytics for learning and analysis
|
|
-- Enables Phase 5.1 (embedding-based KG) and Phase 5.5 (persistent storage)
|
|
|
|
-- KG Executions: Historical record of all agent task executions
|
|
DEFINE TABLE kg_executions SCHEMAFULL;
|
|
|
|
DEFINE FIELD execution_id ON TABLE kg_executions TYPE string ASSERT $value != NONE;
|
|
DEFINE FIELD task_description ON TABLE kg_executions TYPE string ASSERT $value != NONE;
|
|
DEFINE FIELD agent_id ON TABLE kg_executions TYPE string ASSERT $value != NONE;
|
|
DEFINE FIELD outcome ON TABLE kg_executions TYPE string ASSERT $value INSIDE ['success', 'failure'];
|
|
DEFINE FIELD duration_ms ON TABLE kg_executions TYPE int DEFAULT 0;
|
|
DEFINE FIELD input_tokens ON TABLE kg_executions TYPE int DEFAULT 0;
|
|
DEFINE FIELD output_tokens ON TABLE kg_executions TYPE int DEFAULT 0;
|
|
DEFINE FIELD task_type ON TABLE kg_executions TYPE string DEFAULT "general";
|
|
DEFINE FIELD error_message ON TABLE kg_executions TYPE option<string>;
|
|
DEFINE FIELD solution ON TABLE kg_executions TYPE option<string>;
|
|
DEFINE FIELD embedding ON TABLE kg_executions TYPE array<f32>;
|
|
DEFINE FIELD executed_at ON TABLE kg_executions TYPE datetime DEFAULT time::now();
|
|
DEFINE FIELD created_at ON TABLE kg_executions TYPE datetime DEFAULT time::now();
|
|
|
|
DEFINE INDEX idx_kg_executions_agent ON TABLE kg_executions COLUMNS agent_id;
|
|
DEFINE INDEX idx_kg_executions_task_type ON TABLE kg_executions COLUMNS task_type;
|
|
DEFINE INDEX idx_kg_executions_outcome ON TABLE kg_executions COLUMNS outcome;
|
|
DEFINE INDEX idx_kg_executions_executed_at ON TABLE kg_executions COLUMNS executed_at;
|
|
DEFINE INDEX idx_kg_executions_agent_type ON TABLE kg_executions COLUMNS agent_id, task_type;
|
|
DEFINE INDEX idx_kg_executions_task_outcome ON TABLE kg_executions COLUMNS task_description, outcome;
|
|
|
|
-- Analytics Events: Aggregated metrics for trending and analysis
|
|
DEFINE TABLE analytics_events SCHEMAFULL;
|
|
|
|
DEFINE FIELD event_id ON TABLE analytics_events TYPE string ASSERT $value != NONE;
|
|
DEFINE FIELD event_type ON TABLE analytics_events TYPE string ASSERT $value != NONE;
|
|
DEFINE FIELD agent_id ON TABLE analytics_events TYPE string ASSERT $value != NONE;
|
|
DEFINE FIELD metric_name ON TABLE analytics_events TYPE string;
|
|
DEFINE FIELD metric_value ON TABLE analytics_events TYPE float;
|
|
DEFINE FIELD task_type ON TABLE analytics_events TYPE option<string>;
|
|
DEFINE FIELD tags ON TABLE analytics_events TYPE array<string> DEFAULT [];
|
|
DEFINE FIELD recorded_at ON TABLE analytics_events TYPE datetime DEFAULT time::now();
|
|
DEFINE FIELD created_at ON TABLE analytics_events TYPE datetime DEFAULT time::now();
|
|
|
|
DEFINE INDEX idx_analytics_agent ON TABLE analytics_events COLUMNS agent_id;
|
|
DEFINE INDEX idx_analytics_event_type ON TABLE analytics_events COLUMNS event_type;
|
|
DEFINE INDEX idx_analytics_recorded_at ON TABLE analytics_events COLUMNS recorded_at;
|
|
DEFINE INDEX idx_analytics_agent_type ON TABLE analytics_events COLUMNS agent_id, event_type;
|
|
|
|
-- View: Success rate by agent (for analytics)
|
|
DEFINE VIEW agent_success_rate AS
|
|
SELECT
|
|
agent_id,
|
|
math::round(count(SELECT success FROM (
|
|
SELECT outcome = 'success' AS success FROM kg_executions WHERE outcome = 'success'
|
|
)) * 100.0 / count(*), 2) AS success_rate_percent,
|
|
count(*) AS total_executions,
|
|
math::avg(duration_ms) AS avg_duration_ms
|
|
FROM kg_executions
|
|
GROUP BY agent_id;
|
|
|
|
-- View: Task type distribution
|
|
DEFINE VIEW task_type_distribution AS
|
|
SELECT
|
|
task_type,
|
|
count(*) AS execution_count,
|
|
math::round(count(SELECT success FROM (
|
|
SELECT outcome = 'success' AS success FROM kg_executions WHERE outcome = 'success'
|
|
)) * 100.0 / count(*), 2) AS success_rate_percent
|
|
FROM kg_executions
|
|
GROUP BY task_type;
|