Re:Some argue that synthetic data can make AI systems better
by redytedy on Saturday February 19, 2022 @01:15AM
(#30394818)
The tricky part is that the simulator itself may not have easy to understand rules. Waymo has a Neurips talk about training world agent models that are used for car behavior in the simulation itself. Trying to make world agents that are indistinguishable from real-world vehicle behavior (e.g. minimizing jenson-shannon curvature entropy) is a completely different task than training a model to safely transport you somewhere.
Re:Some argue that synthetic data can make AI systems better
by taeric on Saturday February 19, 2022 @11:10AM
(#30398127)
Right. That was my angle. Sometimes, they are ready to see after the fact. But, just like bias can enter into data collection, so too can it enter simulation.
Re:Some argue that synthetic data can make AI systems better
by csee on Saturday February 19, 2022 @03:05AM
(#30395270)
Yes but isn't the optimal policy novel information?
Re:Some argue that synthetic data can make AI systems better
by taeric on Saturday February 19, 2022 @11:10AM
(#30398132)
But it's it free of bias and a good execution policy in the real world?
Re:Some argue that synthetic data can make AI systems better
by csee on Saturday February 19, 2022 @01:37PM
(#30399327)
That probably depends on how realistic the simulator is. Your trade off is more bias (due to simulation inaccuracy) in exchange for unlimited data (and hence less variance).
Re:Some argue that synthetic data can make AI systems better
by taeric on Saturday February 19, 2022 @02:05PM
(#30399570)
Agreed. I could see a good case of training heavily on simulated data, verify/validate on collected data.
My question is ultimately how this really helps with making the system ethical. Just moved the bias from collection to simulation. And... I can't see simulation being less impacted with bias.
This is really no different than it picking on the rules embedded in the data gathering of real world data. Any implicit and hidden decisions in that space would be expected to find their ways into the ML.