Karen Liu: The New Role of Physics Simulation in AI

When it comes to tasks involving physical human-robot interaction, learning from physically simulated humans and environments enables robots to safely learn from failure without putting real people at risk. However, presently the simulation tool is not very effective, as many designs produced in simulation fail to deliver in the real world—the so-called sim-to-reality gap. The sim-to-real problem is further complicated by deploying policies to physically interact with people in the real world.

While most recent work on sim-to-real transfer problems focused on improving control policies, Stanford professor Karen Lui's work shows that one can overcome the sim-to-real gap by improving physics simulation processes. In this HAI seminar, she presents her recent work on creating “learnable” physics engines along with efficient techniques for training them. She also reports on the current progress on sim-to-real transfer with humans in the environment.