Learn Before
In an autonomous agent, a procedural-memory gap begins with an inability to correctly diagnose the operational failure during its initial occurrence.
0
1
Tags
Prep Sessions
Long-Horizon Agent Reliability: Stateful Scaffolding and Runtime Verification @ University of Michigan - Ann Arbor
Ch.1 Foundations and Operational Challenges - Long-Horizon Agent Reliability: Stateful Scaffolding and Runtime Verification @ University of Michigan - Ann Arbor
Harness Scaling Foundations and Control Design Space - Long-Horizon Agent Reliability: Stateful Scaffolding and Runtime Verification @ University of Michigan - Ann Arbor
Operational Gaps in Long-Horizon Agent Execution - Long-Horizon Agent Reliability: Stateful Scaffolding and Runtime Verification @ University of Michigan - Ann Arbor
Ch.3 State Persistence and Continuous Optimization - Long-Horizon Agent Reliability: Stateful Scaffolding and Runtime Verification @ University of Michigan - Ann Arbor
Failure-Driven Harness Optimization and Procedural Memory - Long-Horizon Agent Reliability: Stateful Scaffolding and Runtime Verification @ University of Michigan - Ann Arbor
Related
An agent that has the correct procedure active during an execution but fails to complete or follow it is exhibiting a procedural-compliance gap rather than a procedural-memory gap.
Why is a procedural-memory gap classified specifically as a longitudinal failure mode in autonomous agents?
In an autonomous agent, a procedural-memory gap begins with an inability to correctly diagnose the operational failure during its initial occurrence.
How does the prior acquisition of knowledge differ between an epistemic gap and a procedural-memory gap when an autonomous agent encounters an operational risk?
Across separate task runs, an autonomous agent successfully identifies the cause of an operational issue during the first run. However, when the exact same risk appears in a subsequent run, the agent fails to retain or apply the lesson it previously established. What type of failure mode does this represent?
In the context of agent failure modes, how is an epistemic gap defined, and at what specific point does it manifest?