- 0
- 0
- 约6.37千字
- 约 11页
- 2026-06-01 发布于江苏
- 举报
PrinciplesofDeepRL
DavidSilver
-2-
Principle#1:EvaluationDrivesProgress
Objective,quantitativeevaluationdrivesprogress:
●Thechoiceofevaluationmetricdeterminesthedirectionofprogress
●Arguablythemostimportantsingledecisioninthecourseofaproject
Leaderboard-drivenresearch:
●Besuretheevaluationmetriccorrespondscloselytotheendgoal
●Avoidsubjectiveevaluation(e.g.humaninspection)
Hypothesis-d
您可能关注的文档
- 人工智能核心550篇论文.机器学习 tidy-data.pdf
- 人工智能核心550篇论文.姿态估计 1904.04514v1.pdf
- 人工智能驾驶电子版 技术文档0364.pdf
- 人工智能驾驶电子版 精选论文04.The Microsoft 2016 Conversational Speech Recognition System.pdf
- 人工智能驾驶电子版 精选论文06.Accelerating very deep convolutional networks for classification and detection.pdf
- 人工智能驾驶电子版 精选论文10. DoReFa-Net_ Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients.pdf
- 人工智能驾驶电子版 精选论文Integration Challenges of Facilities-Layer DCC for Heterogeneous V2X Services.pdf
- 直角三角形中圆位置关系及应用实例分析.pdf
- 函数零点题型拓展与方法归纳.pdf
- Lufkin 7.1 4寸-184MM 24齿木工圆锯产品信息.pdf
原创力文档

文档评论(0)