(new article with draft notes. Not full sentences but keypoints are provided about the subject of learning from demonstration and chatbots) |
m (minor typo correction) |
||
Line 5: | Line 5: | ||
* a robot can replace a human worker only, if he has human level AI |
* a robot can replace a human worker only, if he has human level AI |
||
* sliding mode = model predictive control |
* sliding mode = model predictive control |
||
− | * instead of calculating the complete forward model, only rewards are measured <ref>Gaina, Raluca D., Simon M. Lucas, and Diego Pérez |
+ | * instead of calculating the complete forward model, only rewards are measured <ref>Gaina, Raluca D., Simon M. Lucas, and Diego Pérez-Liébana. "Tackling sparse rewards in real-time games with statistical forward planning methods." Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 33. 2019.</ref> |
* Phantom Auto: teleoperated car |
* Phantom Auto: teleoperated car |
||
* productivity paradox "teleoperation" |
* productivity paradox "teleoperation" |
||
* Model Predictive Control for time delay teleoperation |
* Model Predictive Control for time delay teleoperation |
||
* robot programming != teleoperation |
* robot programming != teleoperation |
||
− | * programming by demonstration |
+ | * programming by demonstration -> trajectory planner |
* “multiple demonstrations” trajectory database |
* “multiple demonstrations” trajectory database |
||
− | * Multiple task learning, task*parameterized Gaussian mixture model (TP |
+ | * Multiple task learning, task*parameterized Gaussian mixture model (TP GMM) |
* parameterized trajectories |
* parameterized trajectories |
||
* "Generative Adversarial Networks" "learning from demonstration" |
* "Generative Adversarial Networks" "learning from demonstration" |
||
Line 19: | Line 19: | ||
==Chatbots== |
==Chatbots== |
||
− | * chatbots: AIML, deeplearning. Chatbot types: task |
+ | * chatbots: AIML, deeplearning. Chatbot types: task-oriented/goal-oriented (restaurant reservation) or open-domain. <ref>paper “Richárd Csáky: Deep Learning Based Chatbot Models, 2017”</ref> |
* AIML = chatbot description |
* AIML = chatbot description |
||
* QAKIS Question Answering grounded |
* QAKIS Question Answering grounded |
||
* chatbots are working with hypothesis anchoring |
* chatbots are working with hypothesis anchoring |
||
* creating textadventures instead of chatbots |
* creating textadventures instead of chatbots |
||
− | * It seems that automatic evaluation of a dialoque is the weak point of current chatbots. Perhaps this explains why GPT2 is using a co |
+ | * It seems that automatic evaluation of a dialoque is the weak point of current chatbots. Perhaps this explains why GPT2 is using a co-evolution strategy in which the generation of dialogues and the evaluation of speech is trained separate. |
* AIML describes a chatbot corpus |
* AIML describes a chatbot corpus |
||
* chatbot languages: ChatScript (2011) very powerful but huge size to download, AIML (2001) outdated |
* chatbot languages: ChatScript (2011) very powerful but huge size to download, AIML (2001) outdated |
||
Line 31: | Line 31: | ||
==References== |
==References== |
||
<References/> |
<References/> |
||
− | |||
[[Category:2020]] |
[[Category:2020]] |
Latest revision as of 12:49, 17 February 2020
2020-02-17
Learning from demonstration
- a robot can replace a human worker only, if he has human level AI
- sliding mode = model predictive control
- instead of calculating the complete forward model, only rewards are measured [1]
- Phantom Auto: teleoperated car
- productivity paradox "teleoperation"
- Model Predictive Control for time delay teleoperation
- robot programming != teleoperation
- programming by demonstration -> trajectory planner
- “multiple demonstrations” trajectory database
- Multiple task learning, task*parameterized Gaussian mixture model (TP GMM)
- parameterized trajectories
- "Generative Adversarial Networks" "learning from demonstration"
- learning from demonstration with spline interpolation
Chatbots
- chatbots: AIML, deeplearning. Chatbot types: task-oriented/goal-oriented (restaurant reservation) or open-domain. [2]
- AIML = chatbot description
- QAKIS Question Answering grounded
- chatbots are working with hypothesis anchoring
- creating textadventures instead of chatbots
- It seems that automatic evaluation of a dialoque is the weak point of current chatbots. Perhaps this explains why GPT2 is using a co-evolution strategy in which the generation of dialogues and the evaluation of speech is trained separate.
- AIML describes a chatbot corpus
- chatbot languages: ChatScript (2011) very powerful but huge size to download, AIML (2001) outdated
- pddl chatbot -> dialoque plan: The agent comes up with a plan to solve its task based on the domain description provided.