We built the Aiseedo engine to make cutting edge Machine Learning available to everyone and enable the creation of smart, adaptive and engaging products.
This demo gives you a peek at what the Aiseedo engine can do.
The Aiseedo engine controls the cookie monster using Closed Loop Analytics to guide it through its world by taking actions (back-/forward, left/right) to maximise its rewards (cookies).
When it was first started the engine had no information about its environment. It did not know the configuration, the rules or the goal of the game, it had to learn it all. More importantly, we didn’t have to tell it about the messages it would receive, nor configure the fields. We simply told it it to pick actions to maximise the score, and let it go from there.
The engine is leveraging an innovative combination of several machine learning techniques (feed forward neural nets, recurrent neural nets, reinforcement learning and data fusion) to extract patterns from the message streams and build a holistic predictive model of the cookie monster's world.
Data fusion is the integration of multiple varied sources of information to reduce noise and improve understanding. Aiseedo was designed to merge messages from different sources on different timescales, for example the cookie monster's sense of touch, movement, and sight. Touch events happen irregularly, movement messages depend on the input of the Aiseedo engine or of the player, while the cookie monster gets continued sight of its environment.
Closed Loop Analytics are an advanced Machine Learning technology that relies on a live feedback loop between a Machine Learning engine and the real world. The Aiseedo engine suggests what it believes to be the best action for the monster to be happy (obtain rewards). The engine then uses the feedback from testing these suggestions to learn and correct its model of the world.
Until now the approach in most machine learning applications is to dictate the nature of the solution, and train the model to fill in the details. We believe that setting a goal and relying on machine learning to solve the problem will enable more reslilient and flexible solutions. It is an exciting step towards Machine Intelligence.
In this demo the cookie monster is guided to optimize its actions by imagining only a couple of seconds into the future… for now :)
To give feedback contact us with the link below.