What is the revenue optimal selling mechanism for two buyers with i.i.d. uniform valuations over two items?

Mechanism design is very hard. One of the most exciting lines of work for me is towards automating mechanism design via deep learning, sprouting from this. There is a very compeling talk by David Parkes on this line of work:

Daniel Reusche and I wrote up a short note on a ongoing line of work to use differential privacy in the learning of the mechanism, building up on the classic work of Nissim et al. Recent work on improving the bounds of DP we hope will help in giving bounds on the composed mechanisms. Daniel and Michael Curry also have a interesting blog post describing a GAN based approach to mechanism learning.

I am very keen in bringing this to the attention of the ethereum and those exploring novel public good funding mechanisms. Those interested in exploring potential field experiments that can take place, where the computational contract (perhaps on the zkEVM alpha) learns the incentives it can provide to maximize notions of social welfare or profit.

Using regret minimization for aproximately truthful mechanism design was used in the use it to design strategy-proof, multi-facility mechanisms that minimize expected social cost.

https://arxiv.org/pdf/1004.2888.pdf https://arxiv.org/pdf/2205.03470.pdf