Demo of Virtual Dog Learning to Play Fetch via Imitation and Reinforcement

[XxJBieberxXLuverxX]

what game is this?



[lolamaya00]

what is this game???

[JermskiB]

Great work! I am hoping to leverage the NLP pipeline in OpenCog to create a domain specific chat bot for knowledge delivery.

[MrSamirAraujo]

@JermskiB Yeah. It would be great to see a demo using the NLP module in a chatter bot! :)

[MrSamirAraujo]

@JermskiB it was added, some time ago, a basic and experimental support for chatter bots into the OpenCog virtual agent module. Please, look for "Dialog Controller" in the OpenCog wiki. There you can find some information of how to integrate your own chatter bot system into OpenCog. Ah, there is also a basic chatter bot implementation outside the virtual agent module. It can be integrated to a IRC channel. You can check it out working at our IRC channel: #opencog on Freenode

[randomchickgirl]

What game is this???

[MrSamirAraujo]

@randomchickgirl Hi there! In fact, it isn't a game. It is just a demonstration of OpenCog controlling a virtual agent. OpenCog is a set of Artificial Intelligence tools that can be used, for instance, to create intelligent agents, like the dog shown at this video. We've used Multiverse (a virtual world manager tool) to manage and render the environment. If you want more information about OpenCog, please click on the link informed at the description of this video.