My robot doesn’t use machine learning, planning programs, heuristic searches, graph searches, predicate calculus, common sense systems, rule-based systems, induction/deduction, neural networks, genetic programming, the triple-tower architect, language parsers, grammar rules, Bayesian’s network for decision making, Bayesian’s probability theories, decision trees, data mining, etc, etc.
This robot learns information simply by turning on its 5 senses (sight, sound, taste, touch and smell) and letting it learn everything. Teachers in school teach the robot a variety of skills: decision making, solving problems, generating logic, interpreting complex instructions, understanding natural language, doing multiple tasks, adapting to a changing environment, learning a skill, learning different subject matters, learning to fly a plane, etc. No expert programmers are ever needed for data input.
This video shows a robot playing castlevania for the Sega Genesis. There are no sound in the video because i wanted to show the viewers what the robot is thinking when playing a side-scrolling game. At the beginning of the game, the robot decides to play castlevania. Next, knowledge about playing videogames and playing side-scrolling games pour into primarily 4 containers: task container, rules container, planning container, and identity container. These containers contain things like tasks to do, rules to follow and strategies to use in order to play the game. The flashing text and the freeze frames are internal instructions of the robot. The robot’s thoughts include things like: do tasks, make decisions, solve problems, retrieve facts, generate common sense and so forth.
In this video, I show how the robot can adapt to a changing environment. There are two parts to this video: 1. navigating in a mirror environment. 2. navigating in a up-side down environment. In both gameplays the robot is using logic and common sense to navigate safely from one location to the next.
The video also shows the robot doing multiple tasks at the same time. Decisions are made based on three tasks: 1. collect options. 2. navigate to the end of level. 3. avoid or kill enemies. The robot will factor in all three tasks and determine what kind of action to take in the future. The robot’s intended goal is to pass level 4 and to have high life meter at the end.
My intended goal of this video is to show people that this robot can adapt to any changing environment. For example, if the US government changes traffic rules: green light is red light and vice versa; and right is left and vice versa, this robot has adaptive behavior that will allow him to change the traffic rules in his brain. This adaptation doesn’t just apply to driving a car, it applies to all job skills, regardless of their complexity.
Another example is driving during seasons. If the environment is icey and wet, the driver has to drive a certain way. Just like in a videogame such as castlevania, the environment might be icey and this changes how the character moves in the game. How the character jumps, walks, and climbs are different in a icey environment. The player has to adapt to the changing environment in order to pass the level.
How does the robot learn to adapt to a changing environment? The answer is through teachers in school. Teachers teach the robot how to adapt to a changing environment. Lessons on observation, rule changes, and common sense decisions are taught by teachers starting from elementary school. As the robot learns more, he forms pathways in memory that can adapt to “any” changing environment, regardless of complexity.
For more information about human level artificial intelligence visit my website:
https://www.humanlevelartificialintelligence.com
Post time: May-02-2017