Virtual assistant A.I. and virtual enviroments

 Beyond second life san-sar open sim.
As I see it.
Some mashup fun put in the piece also.
Linden labs has run a model on a game engine. Yet it fails in some key ways. My first recommendation is to find a good game engine like unreal engine. You end up with more options for building in the end. Unreal engine even has a shop setup already but a bit to high priced for general use and more for moders and game builders ATM. Then go out and find the editing tools you want. Blender is a good free mesh editor, but a lot exist out there and various prices. Just make sure your mesh editor interlinks with your game engine is about all I would stress there. Unity,Unreal,Cryengine are decent engines and unreal even comes ready for A.I. to be plugged into it. Take this as a example.
http://phys.org/news/2012-09-artificially-intelligent-game-bots-turing.html
Some of the better games out there have good A.I. systems in them. Just part of what makes virtual environment A.I.’s and more so story telling crafted to the person and can draw from the resources allowed and mapped out.
Unreal Engine 4 AI Tutorial Introduction and Theory
Basics of AI
Setting Up Advanced AI
Unreal Engine Training Livestream – AI Perception & EQS
Unreal Engine 4 Training – AI Perception & EQS Contd
NVIDIA Opens PhysX Code to UE4 Developers
Vulkan on NVIDIA GPUs presentation, by Piers Daniell (SIGGRAPH 2015, SIG1501)
So a fast summary. You have your game engines and graphics card cores to run and train a deep learning network. With multicard SLI setups. Some of the extra cores in the GPU can be used for A.I. along with the CPU cores. The best balancing will vary from machine to machine and library used. Now think of the type of A.I.’s you would put in a virtual environment. From the human with scripted parts or deep precise indexes to deep learning networks that customize the playing experience to the individual. From a personal assistant like siri/cortiana to A.I. that knows every thing about you and can customize the learning experience or play to the individual person. Scenes rendered off of mood from EEG pickups or music and A.I. drawing from many story lines to create a new story experience to play in. Schools and workplaces able to overlay 3D augmented reality with rich text for details and such.
Roger Waters – Welcome to the machine
List of Best Chatbots to Converse with
Computer program to take on world’s best in Texas Hold ’em
To show that humans can be fooled with some small and what I would say is scripting to a A.I. and the light talk everyday so to say. The deeper thought and meaning linking would have to be done with a IBM like probabilistic network and the ability to tie into the lower end scripted words or avatar movements. Some can be thought of as reactive surroundings and natural reactions with in a surrounding. Like fast  reflex reactions are to some prescripting, deep thought patterns and weighted decisions will simulate or be like deeper thought or deep learning networks.
Pink Floyd – Welcome To The Machine (Subtitulos en Español)
Mitsuku chatbot has good answers for the Loebner Prize
Anticipation
Robots Hallucinate Humans to Aid in Object Recognition
Human disguise
Yes, androids do dream of electric sheep
Dynamix II – Machine Planet
Queensryche – Welcome to the Machine
The thing they also seem to push to the side. Yet it also seems to be in front of them so to say. Not just a persons AVI/toon should be transferable threw game engines and into hub’s with a UUID system that branches across engines and companies. But they also need to be able to let 1 or 2 A.I.’s work across platforms as well. In this way the A.I. can monitor the individual across game types and skill levels and knowledge base as well as health and mental fitness. Gamifying in a way, though more a constructive way to have A.I. help people across virtual environments and game engines.
Welcome To The Machine – Pink Floyd
Clueless Program Plays Atari Games, Learns to Win
PhD student creates AI machine that can write video games
Computer writes its own fables
Team uses artificial intelligence to crowdsource interactive fiction
Diablo series showing procedural algorithmic map generation.
Roger Waters – Welcome To The Machine ( In the Flesh )
Real-Time EEG-Based Happiness Detection System
Game on: Sports Analytics Conference produces a growing avalanche of quantitative research
UDK – Kinect Speech Capture Interface
Welcome to the Machine – le Macaque Cover
A Neural Network on GPU
Restricted Boltzmann Machines and Deep Belief Networks on multi-core processors
Pink Floyd – Welcome to the Machine
12. Learning: Neural Nets, Back Propagation
Introduction to Deep Learning with Python
Nice GUI
Neural Network Simulator
Getting Started with Neural Network Toolbox
Intelligence and Machines: Creating Intelligent Machines by Modeling the Brain with Jeff Hawkins
Architects of the Mind: A Blueprint for the Human Brain Full Program
“Welcome to the Machine” performed by Brit Floyd – the Pink Floyd tribute show
The Australian Pink Floyd Show Documentary – Welcome To The Machine
Pink Floyd – Welcome to The Machine
The main companies with the ability to do this are the big ones. Facebook is playing around with some of the tech as well as Microsoft and Google and Steam and Apple. They all have headsets out of types or are in the process of putting them out and lack various parts. From EEG pickup to form and function worked out better in some cases. They also have the research into A.I. and OS’s in a way. Unix will have it’s own ways for the supported game engines, and side development of a type. A virtual assistant that knows everything about the individual person or best trained over a lifetime in a way. A A.I. assistant that can travel from game engine to game engine and some in limited ways. A UUID system that goes from game engine to game engine with mesh AVI/toon limited to some games also. A gamer/Dev hub’s ( mainly gamer/AVI hub ) For meeting friends or even new people in game oriented hubs. Client side A.I. that travels with the person and companies offering cloud point to point and data mining the A.I. does for the person. ( Backend A.I. work ) Steam and Microsoft seem to be setup best for making a hub and allowing some cross platform mixing of AVI to game’s and game engines. Unreal engine is a free engine to start on and people will need or want a client side or solo server they can alter in any way they like. In a way that’s why I hype them so much and how good they have been in the past to the moder community. With Mocap and VR/AR available and good A.I. and scripting of agents in game engines or background A.I. controlling the game flow and direction. The possibilities of engaging game play with feedback and story algorithms customizing the players experience sounds nice. At some point they will need to put a tactile layer in with the mesh layering for when that comes of age so to say. With so few tactile feedback systems out there today, it is a lower priority.
Shadows Fall- Welcome To The Machine
I guess the main things I would stress here are. Scripting the smaller lighter chat/voice over and caricature animation’s to match if any. Look at the existing videogames out there and the small chat they have and how that can be expanded on for types or personalities even. Leave deeper context to IBM Watson like probabilistic deep mining and weighting of trees for deeper context mapping to a A.I. agent. ( harder problem ) Mimic like algorithm for text to voice AVI mouth animations, and persona like animations to go with the personalized A.I. agent. The hard problem is deep context and in a way they have solved part of that already. The other parts should come in time. UUID integrated AVI and AI bridge and standards would be nice to see. If they get it right it will also be medical data along with geospatial or even family history that goes along with the A.I.. The way I see it, people can cobble together large chunks of what they would want already. The first step is good mimicry in A.I. and virtual environments. The next step is small things most do not think of as the talk or react. ( part of personalities even ) and the deeper thinking or thought lines and construction of such to start to give a type of depth. The other thing I would like to say is look at it from the perspective of what does the person need from the A.I. agent. From monitoring to structures of context that the A.I. can feedback to the user. Before you try to go smarter than human right off the bat. Just conquer parts at a time and build up or cobble together other work out there already done in the field.
Pink Floyd – Welcome To The Machine (Person Of Interest)
Last note. On A.I. avatars and personas. Your A.I. can have a persona of types just from avatar movements. AO’s ( Animation Overrides ) show this to a degree and animation packs threw other vendors. The small things like that and simple A.I. chat dialog that fits to area or person or query or such. Part of the trick is building up on that preconceptual illusion or trained neural chain that is part of the small things. From movement to speech to path solving and mimicking a human in ways. From Watson to bots that hallucinate humans to anticipate some things, to siri/Cortina like chat bots with deep context engines running in the background and referenced for the individual or companion A.I. linked to a individual. High Definition poly models and high Def. textures with current technology. You can build a decent system with A.I. agent and a lot of work. VR/AR will probably get A.I. assistants of types along with game engine’s and as they evolve. After all some games come with companion A.I. in the game and done up in a basic way. This seems the natural extension of in game A.I. companion into a A.I. companion that travels with you and in ways monitors you. Work on the parts you see in mimicking human like or a smarter stupid A.I. is another way to put it. Cobble enough parts together and you will have smarter than human level and of a type. Fleshing out other parts will happen as time goes by. ( personality traits, Deep learning neural nets, algorithms, voice packs or expansion packs and so on.
Pink Floyd – Welcome To The Machine – Paris (1977) Legendado
Welcome to the machine (ZeroPointOne remix)
Pink Floyd / Welcome To The Machine / Oforia Mix
Misc mashup
Tinkering with evolution: Ecological implications of modular software networks
Welcome To The Machine *Pink Floyd Cover • Square Tube, Round Heart Mix • Eric Clayton Heard
Welcome to the Machine – Trailer
Steve Aoki & Angger Dimas ft. My Name is Kay – Singularity
Pink Floyd – Welcome To The Machine
Building The Machine (Remix) – Post Human Era
Matrix- Welcome to the Machine
Advertisements

About laborious

Interests in science, art, math, theory, game engines, future technology.
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s