Session
High Fidelity Simulations: The Critical Path to Real-World AI
with Ashish Kapoor
Developing and testing real-world AI is an expensive and time consuming process. Machine Learning is one of the key component that enables these systems to operate under uncertainty. However, in order to utilize recent advances in machine intelligence and deep learning we need to collect a large amount of annotated training data in a variety of conditions and environments. Also, such data driven system are far from perfect and can result in failure cases that can jeopardize safety. In this session we will explore how high fidelity simulations can help us alleviate some of these problems. We will discuss how such near-realistic simulations can help not only with gathering training but also can be embedded in imitation-learning or reinforcement learning loops in order to improve sample complexity. Our discussion will center around AirSim, an open-source simulator built on Unreal Engine that offers physically and visually realistic simulations.
Developing and testing real-world AI is an expensive and time consuming process. Machine Learning is one of the key component that enables these systems to operate under uncertainty. However, in order to utilize recent advances in machine intelligence and deep learning we need to collect a large amount of annotated training data in a variety of conditions and environments. Also, such data driven system are far from perfect and can result in failure cases that can jeopardize safety. In this session we will explore how high fidelity simulations can help us alleviate some of these problems. We will discuss how such near-realistic simulations can help not only with gathering training but also can be embedded in imitation-learning or reinforcement learning loops in order to improve sample complexity. Our discussion will center around AirSim, an open-source simulator built on Unreal Engine that offers physically and visually realistic simulations.
Have feedback? Submit an issue here.