Unverified50% confidenceFactExact time
In self-distillation, if the teacher is an EMA or synchronous variant of the student policy, each update of the student changes the teacher, causing the target signal to drift and turning training into chasing a moving target.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOffline distillation trains the student using a static dataset pre-generated by the teacher model, which leads to the problem of covariate shift.72% similarUnverifiedThe paper observes that supervision signals from a fixed teacher become increasingly less worthwhile as the student model's capabilities improve72% similarVerifiedOnline policy distillation (OPD) has the student model generate its own training responses, which are then supervised by the teacher model.70% similarUnverified教师-学生蒸馏框架先训练可访问完整环境信息的教师策略，再蒸馏到仅依赖机载传感器的学生策略中以实现真实硬件部署65% similarUnverified自适应课程（adaptive curriculum）用智能体当前表现作为反馈信号驱动课程进度，常见实现包括基于成功率阈值的进阶、基于策略梯度信号的难度调节，以及利用进步速度作为代理指标65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1037322API
curl https://kongchang.com/api/v1/knowledge/claims/1037322MCP
get_claim(id=1037322)