Unverified50% confidenceFactExact time
In reinforcement learning, the reward signal is a proxy for the designer's intent, not the intent itself.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedReward Shaping guides the learning direction by designing intermediate rewards, such as setting each point of damage dealt to a boss as an intermediate reward.78% similarUnverifiedThe greatest risk of self-improvement comes from the reward signal itself: if the reward is improperly defined, the agent will optimize evaluation metrics rather than the true objective (the reinforcement learning version of Goodhart's Law).78% similarUnverified高度依赖意图信号可能导致销售资源被误导向研究型访客而非真实决策者，因为意图信号反映的是行为代理指标而非购买意愿的直接证明71% similarUnverifiedTRACER designs hierarchical rewards that provide reward signals at both the outcome level and the trajectory level, combined with bias-aware advantage modulation to mitigate reward sparsity and credit assignment problems.69% similarUnverifiedThe streak mechanism originates from the theory of operant conditioning in behavioral psychology, reinforcing habitual user engagement through variable rewards.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1007090API
curl https://kongchang.com/api/v1/knowledge/claims/1007090MCP
get_claim(id=1007090)