A nice review of our recent work by . "in my view, this is a landmark new idea, which brings the power of modern machine-learning to bear on a central, intractable statistical problem in particle physics."
I wrote something about and and other's very cool new papers on likelihoods for particle physics. (Warning: it's a bit technical)
To find new particles in LHC data, you need a neural network that can learn how to do hard integrals.
In gives a nice short summary of some exciting recent work done by Johann Brehmer, , , and on simulation-based inference for effective field theories in particle physics!
Viewpoint: Fast-Forwarding the Search for New Particles - Physics