[KongchangAI]
Unverified50% confidenceFactExact time

DIAYN训练一个判别器q_φ(z|s),根据状态s推断技能编号z,策略的内在奖励定义为判别器的对数概率log q_φ(z|s)

1
Sources
50%
Confidence
Long-term
Relevance
10/6/2026
First Seen

Sources

SAC训练崩溃之谜:策略熵坍塌与DIAYN调试指南

Related Entities

Cite This Claim

Stable URI
https://kongchang.com/claim/982250
API
curl https://kongchang.com/api/v1/knowledge/claims/982250
MCP
get_claim(id=982250)