Do They Live in a Simulation? Training Models for Dynamic Environments
In this third and final presentation, we will see that synthetic data applications are not limited to static situations.
Please, check the slides here.
Do we live in a simulation?
Computer simulations are widely used and consolidated resources to make estimates and predictions of processes that are difficult or expensive to investigate directly. However, simulations can be combined with machine learning techniques such as reinforcement learning. In this third and final presentation, we will see that synthetic data applications are not limited to static situations.
A room, a factory and even a city can be simulated to establish some relationship with their physical counterparts, a concept known as Digital Twin. For instance, in April of this year, during the opening of the GPU Technology Conference (GTC), Nvidia’s annual conference, there was an excerpt dedicated to commenting on how BMW, the vehicle manufacturer, has been simulating one of its factories to improve the planning of the its production and train robots to perform tasks.
In addition to providing control over data, simulations are environments in which the training of intelligent agents such as robots and autonomous vehicles can be conducted safely and at a faster pace than in real life. With this promising topic, we close this cycle of lectures. What will be the next frontier?