Agent memory should change only after a pattern persists
Several choices are really the same problem: what to forget, when to accept a fact that conflicts with old memory, how quickly to lower trust in a bad source, and how many answer samples to test in a setup. Each choice asks how fast new information should update the system. If updates happen too quickly, one bad input can corrupt memory.
If updates happen too slowly, the agent misses a real change. At the first moment something looks different, a one-off error and the start of a real change look identical. A safer rule is to wait for the same kind of change to appear across a few supporting samples before updating memory.
On 16 expert-labeled NAB data streams, this -based method always won or matched the simple point-by-point rule for lasting changes. In one server mis case, at the same recall, the simple rule produced 1,181 while the -based method produced none. The tradeoff is that the simple rule can be better for short, temporary spikes.
Key points
- Several decisions reduce to one question: how fast should new information be trusted?
- Updating memory from one odd input can let bad data corrupt later behavior.
- A -based rule waits for repeated supporting samples before changing memory.
- Across 16 labeled NAB data streams, lasting changes were handled better or equally well by the -based method.
- Simple point-by-point rules can still be better when the goal is to catch short temporary spikes.