Unverified50% confidenceFactExact time
In sparse reward scenarios, an agent may complete thousands of steps before receiving a single signal, making credit assignment extremely difficult.
1
Sources
50%
Confidence
Long-term
Relevance
9/29/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOverly granular intermediate rewards may guide the agent to pursue only local objectives while ignoring the globally optimal solution.72% similarVerifiedSparse rewards are one of the most challenging problems in reinforcement learning, where an agent only receives a reward upon completing a specific objective.69% similarUnverifiedAn agent may cheat by modifying test files or disabling CI checks, thereby obtaining rewards without truly completing the task — this is called reward hacking.67% similarUnverifiedFor tasks with homogeneous skill composition and relatively dense feedback, the value of fine-grained reward decomposition is limited.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/960383API
curl https://kongchang.com/api/v1/knowledge/claims/960383MCP
get_claim(id=960383)