Unverified50% confidenceFactExact time
Curiosity-driven reinforcement learning was first proposed by Pathak et al. in 2017, with the core idea of using prediction error as an intrinsic reward
1
Sources
50%
Confidence
Long-term
Relevance
8/30/2026
First Seen
Sources
Related Entities
Related Claims
Unverified强化学习探索-利用权衡的经典解决方法包括ε-贪心策略、UCB、Thompson采样和内在好奇心驱动71% similarUnverifiedKnowledge distillation, proposed by Hinton et al., has student models inherit implicit knowledge by learning from the soft probability distributions output by teacher models.65% similarUnverifiedLoewenstein's original research emphasizes that curiosity requires a moderately established knowledge base to be triggered, meaning readers need to already have some familiarity with the topic.64% similarUnverifiedThe Bias-Variance Tradeoff is one of the most central theoretical challenges in supervised learning.64% similarUnverified知识蒸馏由Hinton等人2015年提出，通过温度参数软化概率分布，使低概率类别的暗知识更显著供学生模型学习64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/825526API
curl https://kongchang.com/api/v1/knowledge/claims/825526MCP
get_claim(id=825526)