Weibull Decay

Mnemo is the first AI memory framework to use a cognitive science-based forgetting model.

Why Forgetting Matters

Other memory systems store everything forever. This leads to:

• Noise accumulation — outdated facts drown relevant ones
• Contradictions — old preferences override new ones
• Slow retrieval — searching through irrelevant memories

Mnemo models human forgetting using the Weibull distribution, the same model used in reliability engineering and cognitive psychology.

The Weibull Function

S(t) = exp(-(t/λ)^β)

Parameter	Meaning
t	Time since memory creation (days)
λ	Scale parameter (derived from half-life)
β	Shape parameter (controls decay curve shape)

Memory Tiers

Different memories decay at different rates, just like in human cognition:

Tier	β	Behavior	Example
Core	0.8	Slow start, then rapid drop	User's name, job, key preferences
Working	1.0	Standard exponential	Recent conversations, tasks
Peripheral	1.3	Fast initial drop, long tail	One-off mentions, trivia

Decay Curves

At half-life = 30 days:

Days	Core (β=0.8)	Working (β=1.0)	Peripheral (β=1.3)
0	100%	100%	100%
15	72%	71%	68%
30	50%	50%	50%
60	28%	25%	21%
90	17%	13%	8%

Tier Promotion

Memories move between tiers based on usage:

• → Core: accessed ≥5 times OR importance ≥0.8
• → Peripheral: not accessed for 90+ days
• Working: everything else (default)

Configuration

const mnemo = await createMnemo({
  embedding: { /* ... */ },
  dbPath: './db',
  decay: {
    recencyHalfLifeDays: 30,   // default: 30
    recencyWeight: 0.5,        // default: 0.5
    frequencyWeight: 0.3,      // default: 0.3
    intrinsicWeight: 0.2,      // default: 0.2
  },
  tier: {
    coreAccessThreshold: 5,
    coreImportanceThreshold: 0.8,
    peripheralAgeDays: 90,
  },
});