Reinforcement

Strengthen memories through explicit reinforcement signals.

Endpoint

POST /api/reinforce

Request

interface ReinforceRequest {
  memory_id: string;
  strength_boost?: number;
  feedback?: 'positive' | 'negative' | 'neutral';
  metadata?: Record<string, any>;
}

Examples

Basic Reinforcement

om = OpenMemory()

# Strengthen a memory
om.reinforce_memory(
    memory_id="mem_abc123",
    strength_boost=0.2
)

User Feedback

# Positive feedback strengthens
om.reinforce_memory(
    memory_id="mem_xyz789",
    feedback="positive",
    metadata={"reason": "user_liked"}
)

# Negative feedback weakens
om.reinforce_memory(
    memory_id="mem_bad456",
    feedback="negative",
    strength_boost=-0.15
)

Spaced Repetition

# Implement spaced repetition
def schedule_review(memory):
    om.reinforce_memory(
        memory_id=memory.id,
        strength_boost=0.25,
        metadata={
            "review_count": memory.review_count + 1,
            "next_review": datetime.now() + timedelta(days=7)
        }
    )

See HMD v2 Specification for decay mechanics and Decay Algorithm for reinforcement strategies.