当前位置:网站首页>Robot reinforcement learning -- first person vs third person
Robot reinforcement learning -- first person vs third person
2022-06-29 04:48:00 【Qianyu QY】
see 17 Robot intensive learning thesis in , When adopting imitation learning , Some papers present data in the first person , Some papers use the third person to demonstrate the data , First person and third person , I think it describes the position of the camera .
stay THIRD-PERSON IMITATION LEARNING The explanation is mentioned in this paper :
first person :the agent is provided with a sequence of states and a specification of the actions that it should have taken.
third person :they observe other humans perform tasks, infer the task, and accomplish the same task themselves.
informally : The first person is to see their own demonstration , Demonstrate observations and... In the trajectory action Same as when testing ; The third person is to watch others demonstrate , Demonstrate observations and... In the trajectory action Need to switch to the first person .
边栏推荐
- JVM memory tuning method
- Research Report on the overall scale, major manufacturers, major regions, products and application segmentation of GPS antenna modules in the global market in 2022
- Collection of common terms used in satellite navigation
- data management plan
- 网络设备设置/取消console口登陆单独密码
- Research Report on the overall scale, major manufacturers, major regions, product and application segmentation of the gsm-gprs-edge module of the Internet of things in the global market in 2022
- The subnet of the pool cannot be overlapped with that of other pools.
- C语言用 printf 打印 《爱心》《火星撞地球》等,不断更新
- Distributed transaction Seata
- 【代码随想录-动态规划】最长公共子序列
猜你喜欢
![[IOT] description of renaming the official account](/img/54/43189f34b81a7441cd46d5c2066970.png)
[IOT] description of renaming the official account "Jianyi commerce" to "product renweipeng"

What is the method of connection query in MySQL

Software architecture experiment summary

What are the MySQL database constraint types
![[Verilog quick start of Niuke network question brushing series] ~ asynchronous reset Series T trigger](/img/e3/cf40fb0131ddeb26bc5beeca03d183.png)
[Verilog quick start of Niuke network question brushing series] ~ asynchronous reset Series T trigger

Proxy mode (proxy)

Technical specifications of Tektronix tds3054b oscilloscope

Continue yesterday's plan: February 16, 2022

Agilent digital multimeter software ns multimeter, real-time data acquisition and automatic data saving

What are the circular statements of MySQL
随机推荐
Gocd is good, but talk about Jenkins
Technical parameters of Tektronix DPO4104 digital fluorescence oscilloscope
【IoT】公众号“简一商业”更名为“产品人卫朋”说明
The last week! Summary of pre competition preparation for digital model American Games
Hot renewal process
LabVIEW显示Unicode字符
Installation and configuration of interrealsense d435i camera driver
如何创建 robots.txt 文件?
笔记本访问台式机的共享磁盘
Software architecture final review summary
On February 15, the market hot money operation and the dragon and tiger list
Research Report on the overall scale, major manufacturers, major regions, products and application segments of semiconductor wafer metal stripping platform in the global market in 2022
【HackTheBox】dancing(SMB)
Gbase 8s must be a DBSA. Solution to failure to start due to path change
Mysql 中的 mvcc原理
February 14 institutional dragon and tiger list and operation of well-known hot money
Collection of common terms used in satellite navigation
innography
Research Report on the overall scale, major manufacturers, major regions, products and applications of electric hydrofoil surfboards in the global market in 2022
Cipher