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

125 lines
4.2 KiB
Rust

use criterion::{black_box, criterion_group, criterion_main, Criterion};
use vapora_knowledge_graph::{TemporalKG, ExecutionRecord};
use chrono::Utc;
async fn setup_kg_with_records(count: usize) -> TemporalKG {
let kg = TemporalKG::new("ws://localhost:8000", "root", "root")
.await
.unwrap();
for i in 0..count {
let record = ExecutionRecord {
id: format!("exec-{}", i),
task_id: format!("task-{}", i),
agent_id: format!("agent-{}", i % 10),
task_type: match i % 3 {
0 => "coding".to_string(),
1 => "analysis".to_string(),
_ => "review".to_string(),
},
description: format!("Execute task {} with description", i),
duration_ms: 1000 + (i as u64 * 100) % 5000,
input_tokens: 100 + (i as u64 * 10),
output_tokens: 200 + (i as u64 * 20),
success: i % 10 != 0,
error: if i % 10 == 0 { Some("timeout".to_string()) } else { None },
solution: Some(format!("Solution for task {}", i)),
root_cause: None,
timestamp: Utc::now(),
};
kg.record_execution(record).await.unwrap();
}
kg
}
fn kg_record_execution(c: &mut Criterion) {
c.bench_function("record_single_execution", |b| {
b.to_async(tokio::runtime::Runtime::new().unwrap())
.iter(|| async {
let kg = TemporalKG::new("ws://localhost:8000", "root", "root")
.await
.unwrap();
let record = ExecutionRecord {
id: "test-exec".to_string(),
task_id: "test-task".to_string(),
agent_id: "test-agent".to_string(),
task_type: "coding".to_string(),
description: "Test execution".to_string(),
duration_ms: 1000,
input_tokens: 100,
output_tokens: 200,
success: true,
error: None,
solution: None,
root_cause: None,
timestamp: Utc::now(),
};
black_box(kg.record_execution(black_box(record)).await)
});
});
}
fn kg_query_similar(c: &mut Criterion) {
c.bench_function("query_similar_tasks_100_records", |b| {
b.to_async(tokio::runtime::Runtime::new().unwrap())
.iter_batched(
|| {
let rt = tokio::runtime::Runtime::new().unwrap();
rt.block_on(setup_kg_with_records(100))
},
|kg| async move {
black_box(
kg.query_similar_tasks(
"coding",
"Write a function for processing data",
)
.await,
)
},
criterion::BatchSize::SmallInput,
);
});
}
fn kg_get_statistics(c: &mut Criterion) {
c.bench_function("get_statistics_1000_records", |b| {
b.to_async(tokio::runtime::Runtime::new().unwrap())
.iter_batched(
|| {
let rt = tokio::runtime::Runtime::new().unwrap();
rt.block_on(setup_kg_with_records(1000))
},
|kg| async move {
black_box(kg.get_statistics().await)
},
criterion::BatchSize::SmallInput,
);
});
}
fn kg_get_agent_profile(c: &mut Criterion) {
c.bench_function("get_agent_profile_500_records", |b| {
b.to_async(tokio::runtime::Runtime::new().unwrap())
.iter_batched(
|| {
let rt = tokio::runtime::Runtime::new().unwrap();
rt.block_on(setup_kg_with_records(500))
},
|kg| async move {
black_box(kg.get_agent_profile("agent-1").await)
},
criterion::BatchSize::SmallInput,
);
});
}
criterion_group!(
benches,
kg_record_execution,
kg_query_similar,
kg_get_statistics,
kg_get_agent_profile
);
criterion_main!(benches);