Typical learning approaches rely on a dataset with outputs curated by humans (an expensive and limiting process). Given a set of inputs and outputs, the traditional learning model (e.g., decision tree, support vector machine, baseline or convolutional neural network) can infer the relationship between the two. Because of human-in-the-loop limitation, a popular state-of-the-art approach is reinforcement learning used by Alphabet DeepMind and other groups; recent success of Alpha Go is built on a reinforcement learning approach. Reinforcement learning is based on learning from examples and collecting feedback through a self-supervising mechanism. Trial-and-error learning is about gathering more data to improve the model, with unsupervised learning referring to learing that does not require human-generated training data. The differentiating factor of reinforcement learning is that there is another independent model that will attempt to evaluate success or failure as a human would.
For example, playing Super Mario: one model is controlling video console inputs and playing while learning from trying again and again. A completely different model is looking at the screen (or other input sources) and deciding whether or not the first model is succeeding or failing (based on points or "game over" message). And that second model represents the "supervision" part. Both models would have to be accurate because if supervision is incorrect then the first model will learn poorly, but they are utterly independent. However, for some tasks it may not be possible to replace human input with a model, particularly as the generating model must learn the distribution of the data (i.e., a description that allows to generate more data).
Once the model is trained, inference tends to be a relatively separate step that uses the previously constructed model. Even in reinforcement learning, after multiple stages of back-and-forth training are applied, a final model is generated which will then produce inferenced output. In our approach we take a more integrated and holistic method, combining inference and training into a continuous process.
The Skive it® team is taking a unique approach that differentiates us from current state-of-the-art AI training strategies. By closely integrating the learning and the inference phases in our model, we can generate results consistent with Online Learning. We're building an entirely new type of AI training system - one that is capable of running effectively on low-power, small memory devices. With Robometrics™, we've created a revolutionary framework that can continually measure / monitor the ‘human-ness’ of robotic systems - all powered by proprietary Deep Learning inference algorithms optimized for the Skive it® chip. This means the framework isn’t reliant on a massive and expensive backend of servers (as is required for many traditional AI training strategies), making it the the perfect training mechanism for the future ‘heart’ of every robot.
Robometricss® can now power your application and systems. For details and pricing, click here.
We are looking for strategic hardware or software partners who can co-invest in Skive it, Inc. If you are interested, please check out our Investor Page and contact us at email@example.com
We are looking for businesses who are interested in partnering with Skive it, Inc. to promote their venues and locations on our web and mobile platforms. If you are interested, please check out our Partner Page and contact us at firstname.lastname@example.org
We are looking for bright, motivated candidates to join the Skive it, Inc. team! Current open positions include: Computer Scientist, Senior Engineer, Software Engineering Intern,and Marketing Intern. If you are interested, please check out our postings on LinkedIn Jobs for more information or contact us at email@example.com