I am in situation where i need to estimate the attenuation of an EM wave . we consider EM wave as collection of photons. These photons when strike with some dust particles they scatter in different direction and for a random path before colliding with next particle. We can estimate the attenuation using monte carlo simulations. I want to ask can applying the deep learning here can improve the estimation of attenuation ?

  • $\begingroup$ Hi @user7341333, welcome to the site. Your question sounds more like a line of research than an actual technical question. In any case, you can look into automatic differentiation variational inference (ADVI), which may help you. PyMC is a Python framework than can do ADVI. $\endgroup$
    – noe
    Mar 26 at 16:16
  • $\begingroup$ Also, you could add more detail about how exactly you use monte carlo simulations to address the problem. $\endgroup$
    – noe
    Mar 26 at 16:44
  • $\begingroup$ Monte Carlo would simulate 10^6 photons then take the ratio of those particles that are not absorbed to calculate the transmittance $\endgroup$ Mar 26 at 20:42
  • $\begingroup$ I was asking about how you "simulate" those photons, i.e. can you express this simulation as a differentiable expression with some stochastic noise sources? Also, please don't add the details as comments; instead, edit your question. $\endgroup$
    – noe
    Mar 26 at 20:49
  • $\begingroup$ @noe i have edited the question $\endgroup$ Mar 27 at 8:28


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.