Luke Hardcastle
@lkhardcastle.bsky.social
Bayesian modelling and computation - postdoc at MRC Biostatistics Unit, University of Cambridge
Reposted by Luke Hardcastle
Usual MCMC algorithms are typically guaranteed to work well when used to sample from target distributions for which
i) mass is reasonably well-concentrated in the centre of the state space, and
ii) the log-density is smooth and of moderate growth.
Outside of this setting, things can go poorly.
i) mass is reasonably well-concentrated in the centre of the state space, and
ii) the log-density is smooth and of moderate growth.
Outside of this setting, things can go poorly.
November 27, 2025 at 10:35 AM
Usual MCMC algorithms are typically guaranteed to work well when used to sample from target distributions for which
i) mass is reasonably well-concentrated in the centre of the state space, and
ii) the log-density is smooth and of moderate growth.
Outside of this setting, things can go poorly.
i) mass is reasonably well-concentrated in the centre of the state space, and
ii) the log-density is smooth and of moderate growth.
Outside of this setting, things can go poorly.
Reposted by Luke Hardcastle
“look how much better my favorite model is compared compared to these other TRASH MODELS😒😒😒 when I specifically chose a DGP that matches my models assumptions and not the others’😌”
November 7, 2025 at 4:13 AM
“look how much better my favorite model is compared compared to these other TRASH MODELS😒😒😒 when I specifically chose a DGP that matches my models assumptions and not the others’😌”
Reposted by Luke Hardcastle
Yes, yes you are. Stop cosplaying "legitimate concerns". When you join a far right rally, organised by well known far right leader Tommy Robinson, with a host of speakers who outright say they are far right, you are a racist. No ifs, buts or maybes. There are no excuses for being there.
September 16, 2025 at 5:21 AM
Yes, yes you are. Stop cosplaying "legitimate concerns". When you join a far right rally, organised by well known far right leader Tommy Robinson, with a host of speakers who outright say they are far right, you are a racist. No ifs, buts or maybes. There are no excuses for being there.
In Newcastle this week attending what should be a very fun workshop on non-reversible sampling! If you are interested in any of the talks (including my own, tomorrow) there are links to join remotely on the conference website!
sites.google.com/view/probai-...
sites.google.com/view/probai-...
Newcastle Non-Reversible Sampling Workshop
Topic: This workshop will bring together researchers at the forefront of modern sampling techniques and related disciplines, to discuss recent advances in the field.
This event is funded by the ProbA...
sites.google.com
September 8, 2025 at 8:53 AM
In Newcastle this week attending what should be a very fun workshop on non-reversible sampling! If you are interested in any of the talks (including my own, tomorrow) there are links to join remotely on the conference website!
sites.google.com/view/probai-...
sites.google.com/view/probai-...
Reposted by Luke Hardcastle
New paper on arXiv! And I think it's a good'un 😄
Meet the new Lattice Random Walk (LRW) discretisation for SDEs. It’s radically different from traditional methods like Euler-Maruyama (EM) in that each iteration can only move in discrete steps {-δₓ, 0, δₓ}.
Meet the new Lattice Random Walk (LRW) discretisation for SDEs. It’s radically different from traditional methods like Euler-Maruyama (EM) in that each iteration can only move in discrete steps {-δₓ, 0, δₓ}.
August 29, 2025 at 3:07 PM
New paper on arXiv! And I think it's a good'un 😄
Meet the new Lattice Random Walk (LRW) discretisation for SDEs. It’s radically different from traditional methods like Euler-Maruyama (EM) in that each iteration can only move in discrete steps {-δₓ, 0, δₓ}.
Meet the new Lattice Random Walk (LRW) discretisation for SDEs. It’s radically different from traditional methods like Euler-Maruyama (EM) in that each iteration can only move in discrete steps {-δₓ, 0, δₓ}.
Reposted by Luke Hardcastle
I am pleased to announce that together with some friends, we are organising a workshop on Non-Reversible MCMC Sampling, taking place at Newcastle University from 8–10 September 2025.
Details on the programme and registration can be found at the workshop website (sites.google.com/view/probai-...).
Details on the programme and registration can be found at the workshop website (sites.google.com/view/probai-...).
July 24, 2025 at 11:28 AM
I am pleased to announce that together with some friends, we are organising a workshop on Non-Reversible MCMC Sampling, taking place at Newcastle University from 8–10 September 2025.
Details on the programme and registration can be found at the workshop website (sites.google.com/view/probai-...).
Details on the programme and registration can be found at the workshop website (sites.google.com/view/probai-...).
Reposted by Luke Hardcastle
Starting from last October, we (@OnlineMCSeminar on Twitter, sites.google.com/view/monte-c...) have been running an online seminar on all aspects of Monte Carlo methods, with about ~30 talks so far. We are currently paused for the summer, expecting to return in September 2025.
May 28, 2025 at 5:03 PM
Starting from last October, we (@OnlineMCSeminar on Twitter, sites.google.com/view/monte-c...) have been running an online seminar on all aspects of Monte Carlo methods, with about ~30 talks so far. We are currently paused for the summer, expecting to return in September 2025.