Unverified50% confidenceFactExact time
CPT utility is a nonlinear functional of the reward distribution, not a linear expectation of reward, which breaks the linear additivity of expectation that traditional optimistic algorithms rely on.
1
Sources
50%
Confidence
Long-term
Relevance
11/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe core of CPT consists of two points: the probability weighting function (assigning excessive weight to small-probability events and insufficient weight to medium- and high-probability events) and the value function (defining gain and loss utility relative to a reference point, with a steeper loss region, and risk aversion in the gain region and risk seeking in the loss region).68% similarUnverifiedPolicy gradient methods (such as PPO) rely on a discount factor γ to exponentially decay future rewards, with a reward t steps away from the current moment retaining only a weight of γ^t.66% similarUnverifiedThe paper attributes the inability of prior methods to distinguish different confidence targets to two problems: the reward signals that distinguish different p values are often lost during optimization, and the raw error penalty grows unboundedly as p approaches 1.64% similarUnverifiedAblation experiments show that increasing example usage does not yield consistent gains, but instead presents a trade-off relationship between dynamics and reward.59% similarUnverifiedFailure-driven PPO fine-tuning may be inferior to plain continuation training, with possible causes including distribution shift, reward-shaping side effects, and curriculum order sensitivity.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1138271API
curl https://kongchang.com/api/v1/knowledge/claims/1138271MCP
get_claim(id=1138271)