A demonstration of Unsupervised AI might be when a robot can think and act responsibly within a given environment, akin to what humans would typically do. In order for machines to replicate human intelligence they require two critical elements, as do humans; time and data. A human exhibits intelligence by first collecting/absorbing data over a period of time. With the integration of data and time markers, any machine that can replicate cognitive processes may exhibit intelligence.
An actual unsupervised learning machine requires negligible human interference. In the same way that a human baby expands its intelligence through observation and guidance, an autonomous machine may evolve simply by observation. Guidance expedites the learning process; however, it might have further implications. You could say that if you aim to prepare a machine for unsupervised learning, you simply need to install an application that will faithfully collect data from an array of integrated sensors. These data will be employed for learning and decision making, and subsequently, these decisions are coordinated back to various motor components without any human interference in the routine. This translates to the negation of tech companies or multiple engineers that might otherwise be necessary to make the machine capable.
The addition of more learning rules to the machine may have the effect of constraining its intelligence via Artificial Narrow Intelligence (ANI). This might reduce the capabilities of an autonomous machine, which may be restricted to doing only limited tasks. Hence, in order to create an unsupervised learning machine, what would be required is a machine and cognitive software that is deployed with a minimum set of rules.
The platform designed by Responsible Machines operates as cognitive software. The platform, when integrated with robots, will collect data from sensors, synthesize data, and output responses to the motor components of the machine, which might comprise arms, legs, or a speech synthesizer.
The logical data model of the platform resembles a neural network, which allows the platform to quickly organize data to learn from incoming data. This data organization is designed to extract/learn and generate weights during the processes of identification, comparison, weight computation, and response initiation; all in real-time. This process loops back as incoming data when the response initiates any kind of feedback, which channels continuous learning.
The platform is equipped with a rules engine that contains a minimum of preset conditions. Additional object level rules may be configured by an engineer if they would like to expedite the learning process. Functions comprising consisting of auto-classification, auto-labeling, triggers, and a rules engine are built into the platform, which is an area that does not function in the same way as humans. Since machines have no capacity for hormonal release to initiate responses, functions such as triggers and preset goals were required to be designed as elements of the platform.
This following example might assist with gaining an understanding of preset goals. Preset goals may be classified as long term and short terms goals, where the long term objective of the machine is to achieve a confirmed state (internal) across all data entities, whereas the short-term objective is to achieve equilibrium (external) at all times.
A detailed article on unsupervised learning machine emulating humans, can be viewed here.
We certainly appreciate any feedback, advice, opinions, or questions that will assist us with gathering diverse perspectives toward improving the capabilities of the platform.