accelerating MCMC

I have recently [well, not so recently!] been asked to write a review paper on ways of accelerating MCMC algorithms for the [review] journal WIREs Computational Statistics and would welcome all suggestions towards the goal of accelerating MCMC algorithms. Besides [and including more on]

• coupling strategies using different kernels and switching between them;
• tempering strategies using flatter or lower dimensional targets as intermediary steps, e.g., à la Neal;
• sequential Monte Carlo with particle systems targeting again flatter or lower dimensional targets and adapting proposals to this effect;
• Hamiltonian MCMC, again with connections to Radford (and more generally ways of avoiding rejections);
• adaptive MCMC, obviously;
• Rao-Blackwellisation, just as obviously (in the sense that increasing the precision in the resulting estimates means less simulations).

Varanasi unholy air

One of my lasting memories of my trip to Varanasi four years ago is the poor quality of the air, with an almost constant fog over the city, fed by open air fires everywhere and aggressive vehicle exhaust, rather than by the few cremation pyres by the Ganges… I read today in The Guardian that the city actually ranks worst in India for its air quality. (I also read in that article that Gwalior had similar issues, although I remember a pleasant walk around the fort there, in the sun. Presumably on one of the few “good” air quality days.) Not that Paris is doing great in the past days, with a whole week of ineffective driving restrictions that left wood heating operating at full blast. I did not feel the air difference while biking, but I presume the impact of the micro-particles central to those pollution alerts is more long-term!

India snapshot (#7)

A meditating monkey on a wall of Amber Fort, heedless of the surrounding crowd…

India snapshot (#6)

Still in snapshot mode, with the view of the procession of elephants carrying tourists up to the magnificent Amber (or Amer) Fort, north of Jaipur. (Getting there in early morning allowed us to avoid the press of visitors and to enjoy the place even more!)

India snapshot (#5)

Another obvious candidate for a snapshot, the iconic Palace of the Winds or Hawa Mahal (हवा महल), Jaipur, in its morning glory before the crowds appeared.