Jesús Pérez d14150da75 feat: Phase 5.3 - Multi-Agent Learning Infrastructure
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.
2026-01-11 13:03:53 +00:00

188 lines
5.9 KiB
Rust

use std::collections::HashMap;
use vapora_llm_router::{BudgetManager, RoleBudget};
fn create_test_budgets() -> HashMap<String, RoleBudget> {
let mut budgets = HashMap::new();
budgets.insert(
"architect".to_string(),
RoleBudget {
role: "architect".to_string(),
monthly_limit_cents: 50000, // $500
weekly_limit_cents: 12500, // $125
fallback_provider: "gemini".to_string(),
alert_threshold: 0.8,
},
);
budgets.insert(
"developer".to_string(),
RoleBudget {
role: "developer".to_string(),
monthly_limit_cents: 30000,
weekly_limit_cents: 7500,
fallback_provider: "ollama".to_string(),
alert_threshold: 0.8,
},
);
budgets
}
#[tokio::test]
async fn test_budget_initialization() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
let status = manager.check_budget("architect").await.unwrap();
assert_eq!(status.role, "architect");
assert_eq!(status.monthly_remaining_cents, 50000);
assert_eq!(status.monthly_utilization, 0.0);
assert!(!status.exceeded);
assert!(!status.near_threshold);
}
#[tokio::test]
async fn test_budget_spending() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
manager.record_spend("developer", 3000).await.unwrap();
let status = manager.check_budget("developer").await.unwrap();
assert_eq!(status.monthly_remaining_cents, 27000);
assert!((status.monthly_utilization - 0.1).abs() < 0.01);
}
#[tokio::test]
async fn test_multiple_spends_accumulate() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
manager.record_spend("developer", 5000).await.unwrap();
manager.record_spend("developer", 3000).await.unwrap();
manager.record_spend("developer", 2000).await.unwrap();
let status = manager.check_budget("developer").await.unwrap();
assert_eq!(status.monthly_remaining_cents, 20000); // 30000 - 10000
}
#[tokio::test]
async fn test_alert_threshold_near() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
// Spend 81% of weekly budget (12500 * 0.81 = 10125) to trigger near_threshold
// This keeps us under both monthly and weekly limits while triggering alert
let spend_amount = (12500.0 * 0.81) as u32; // 10125
manager.record_spend("architect", spend_amount).await.unwrap();
let status = manager.check_budget("architect").await.unwrap();
assert!(!status.exceeded);
assert!(status.near_threshold);
}
#[tokio::test]
async fn test_budget_exceeded() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
// Spend entire monthly budget
manager.record_spend("developer", 30000).await.unwrap();
let status = manager.check_budget("developer").await.unwrap();
assert!(status.exceeded);
assert_eq!(status.monthly_remaining_cents, 0);
}
#[tokio::test]
async fn test_budget_overspend() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
// Spend more than budget (overflow protection)
manager.record_spend("developer", 35000).await.unwrap();
let status = manager.check_budget("developer").await.unwrap();
assert!(status.exceeded);
assert_eq!(status.monthly_remaining_cents, 0); // Saturating subtract
}
#[tokio::test]
async fn test_weekly_budget_independent() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
// Spend 100% of weekly budget but only 25% of monthly
manager.record_spend("developer", 7500).await.unwrap();
let status = manager.check_budget("developer").await.unwrap();
assert_eq!(status.monthly_remaining_cents, 22500);
assert_eq!(status.weekly_remaining_cents, 0);
assert!(status.exceeded); // Both budgets checked
}
#[tokio::test]
async fn test_fallback_provider() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
let fallback_dev = manager.get_fallback_provider("developer").await.unwrap();
assert_eq!(fallback_dev, "ollama");
let fallback_arch = manager.get_fallback_provider("architect").await.unwrap();
assert_eq!(fallback_arch, "gemini");
}
#[tokio::test]
async fn test_unknown_role_error() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
let result = manager.check_budget("unknown").await;
assert!(result.is_err());
let result = manager.record_spend("unknown", 100).await;
assert!(result.is_err());
}
#[tokio::test]
async fn test_get_all_budgets() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
manager.record_spend("architect", 5000).await.unwrap();
manager.record_spend("developer", 3000).await.unwrap();
let all_statuses = manager.get_all_budgets().await;
assert_eq!(all_statuses.len(), 2);
let arch_status = all_statuses
.iter()
.find(|s| s.role == "architect")
.unwrap();
assert_eq!(arch_status.monthly_remaining_cents, 45000);
let dev_status = all_statuses
.iter()
.find(|s| s.role == "developer")
.unwrap();
assert_eq!(dev_status.monthly_remaining_cents, 27000);
}
#[tokio::test]
async fn test_budget_status_comprehensive() {
let budgets = create_test_budgets();
let manager = BudgetManager::new(budgets);
// Spend 6000 cents: keeps us at 12% of monthly and 48% of weekly (both safe)
manager.record_spend("architect", 6000).await.unwrap();
let status = manager.check_budget("architect").await.unwrap();
assert_eq!(status.monthly_remaining_cents, 44000);
assert!((status.monthly_utilization - 0.12).abs() < 0.01);
assert!(!status.exceeded);
assert!(!status.near_threshold); // 12% < 80%
assert_eq!(status.fallback_provider, "gemini");
}