Generative Agents: Recency Score
Beginner
The problem, from first principles
Generative Agents (Park et al., 2023) retrieve memories by a mix of recency, importance and relevance. Recency decays exponentially with time since the memory was last used, so recent experiences come back first.
From theory to code
Implement recency_score(hours, decay), returning decay ** hours.
Constraints
- The default decay is 0.995 per hour.
Hints
Hint 1
Raise the decay factor to the power of the elapsed hours.
Keywords
research-papersagentsmemorysimulation
Python
Run your code to see results
No results yet
Run your code to check it against the examples, or submit to run every test.