Alex Thiery
@alexxthiery.bsky.social
Associate Prof. in ML & Statistics at NUS 🇸🇬
MonteCarlo methods, probabilistic models, Inverse Problems, Optimization
https://alexxthiery.github.io/
MonteCarlo methods, probabilistic models, Inverse Problems, Optimization
https://alexxthiery.github.io/
And a recent very well written review of NS:
"Nested sampling for physical scientists"
arxiv.org/abs/2205.15570
"Nested sampling for physical scientists"
arxiv.org/abs/2205.15570
June 23, 2025 at 11:32 AM
And a recent very well written review of NS:
"Nested sampling for physical scientists"
arxiv.org/abs/2205.15570
"Nested sampling for physical scientists"
arxiv.org/abs/2205.15570
And here is how the geodesic path looks like (again under the Fisher-Rao metric)
June 13, 2025 at 4:30 PM
And here is how the geodesic path looks like (again under the Fisher-Rao metric)
Here's how the gradient flow for minimizing KL(pi, target) looks under the Fisher-Rao metric. I thought some probability mass would be disappearing on the left and appearing on the right (i.e. teleportation), like a geodesic under the same metric, but I was very wrong... What's the right intuition?
June 13, 2025 at 4:29 PM
Here's how the gradient flow for minimizing KL(pi, target) looks under the Fisher-Rao metric. I thought some probability mass would be disappearing on the left and appearing on the right (i.e. teleportation), like a geodesic under the same metric, but I was very wrong... What's the right intuition?
These sparse Gaussian Processes have been around longer than some grad students, but still fun to code! (and today was my first time coding one...)
April 19, 2025 at 3:18 PM
These sparse Gaussian Processes have been around longer than some grad students, but still fun to code! (and today was my first time coding one...)
Today, re-reading a classic.. the 1953 paper that started it all
April 11, 2025 at 9:00 AM
Today, re-reading a classic.. the 1953 paper that started it all
Cute way to upper bound the connective constant of Z^d. For some length L, enumerate {w_1, w_2, ... , w_N} the Self-Avoiding-Walks of size L. An upper bound is given by the largest eigenvalue of the NxN matrix where M_{i,j}=1 iff there is a SAW of size (L+1) that starts with w_i and ends with w_j.
April 1, 2025 at 9:42 AM
Cute way to upper bound the connective constant of Z^d. For some length L, enumerate {w_1, w_2, ... , w_N} the Self-Avoiding-Walks of size L. An upper bound is given by the largest eigenvalue of the NxN matrix where M_{i,j}=1 iff there is a SAW of size (L+1) that starts with w_i and ends with w_j.
Approximating N(L), the number of Self-Avoiding-Walks in Z^2 of length L, is an assignment in my Simulation course this year. The connective constant is:
C = \lim N(L)^1/L ~ 2.638..
Still open-problem to this day: is it true that 1/C equals the zero of the polynomial P(x)=581*x^4 + 7*x^2 - 13 😱
C = \lim N(L)^1/L ~ 2.638..
Still open-problem to this day: is it true that 1/C equals the zero of the polynomial P(x)=581*x^4 + 7*x^2 - 13 😱
April 1, 2025 at 3:19 AM
Approximating N(L), the number of Self-Avoiding-Walks in Z^2 of length L, is an assignment in my Simulation course this year. The connective constant is:
C = \lim N(L)^1/L ~ 2.638..
Still open-problem to this day: is it true that 1/C equals the zero of the polynomial P(x)=581*x^4 + 7*x^2 - 13 😱
C = \lim N(L)^1/L ~ 2.638..
Still open-problem to this day: is it true that 1/C equals the zero of the polynomial P(x)=581*x^4 + 7*x^2 - 13 😱
This fast way of finding the LIS is neat! Just tried to reproduce your nice plot without leaving the phone 😊
chatgpt.com/share/67e8ec...
chatgpt.com/share/67e8ec...
March 30, 2025 at 7:06 AM
This fast way of finding the LIS is neat! Just tried to reproduce your nice plot without leaving the phone 😊
chatgpt.com/share/67e8ec...
chatgpt.com/share/67e8ec...
Sequential Monte Carlo (aka. Particle Socialism?):
"why send one explorer when you can send a whole army of clueless one"
"why send one explorer when you can send a whole army of clueless one"
March 29, 2025 at 8:32 AM
Sequential Monte Carlo (aka. Particle Socialism?):
"why send one explorer when you can send a whole army of clueless one"
"why send one explorer when you can send a whole army of clueless one"
Next week is the MCMC chapter of my simulation course. Asked chatgpt to come up with a funny drawing:
March 29, 2025 at 8:09 AM
Next week is the MCMC chapter of my simulation course. Asked chatgpt to come up with a funny drawing:
I had to google it. Is it really the plot? 😅
January 30, 2025 at 4:17 PM
I had to google it. Is it really the plot? 😅
and another related paper
arxiv.org/abs/1401.3559
arxiv.org/abs/1401.3559
January 29, 2025 at 9:15 AM
and another related paper
arxiv.org/abs/1401.3559
arxiv.org/abs/1401.3559
January 29, 2025 at 8:37 AM
When implementing parallel tempering, it's fashionable to alternate even and odd index temperature swap to try to maximise the inter-temperature movements. But when the temperatures are appropriately tuned, this very new paper by Roberts & Rosenthal shows that the gains are quite modest!
January 29, 2025 at 8:36 AM
When implementing parallel tempering, it's fashionable to alternate even and odd index temperature swap to try to maximise the inter-temperature movements. But when the temperatures are appropriately tuned, this very new paper by Roberts & Rosenthal shows that the gains are quite modest!
Oh! Diaconis was there 😅
January 28, 2025 at 6:40 PM
Oh! Diaconis was there 😅
Asked to the students of my "statistical simulation" class:
In Buffon's experiment where a needle of length L falls on parallel strips (unit width), the needle crosses 2L/π strips on average. To maximize the accuracy of the resulting estimate of π, how should one choose the length of the needle?
In Buffon's experiment where a needle of length L falls on parallel strips (unit width), the needle crosses 2L/π strips on average. To maximize the accuracy of the resulting estimate of π, how should one choose the length of the needle?
January 28, 2025 at 6:15 PM
Asked to the students of my "statistical simulation" class:
In Buffon's experiment where a needle of length L falls on parallel strips (unit width), the needle crosses 2L/π strips on average. To maximize the accuracy of the resulting estimate of π, how should one choose the length of the needle?
In Buffon's experiment where a needle of length L falls on parallel strips (unit width), the needle crosses 2L/π strips on average. To maximize the accuracy of the resulting estimate of π, how should one choose the length of the needle?
Very neat result by Pozza & Zanella: multi-proposal MCMC schemes are basically not worth it! And with GPUs, the gains are at most modest...
arxiv.org/abs/2410.23174
arxiv.org/abs/2410.23174
January 22, 2025 at 2:22 PM
Very neat result by Pozza & Zanella: multi-proposal MCMC schemes are basically not worth it! And with GPUs, the gains are at most modest...
arxiv.org/abs/2410.23174
arxiv.org/abs/2410.23174
One #postdoc position is still available at the National University of Singapore (NUS) to work on sampling, high-dimensional data-assimilation, and diffusion/flow models. Applications are open until the end of January. Details:
alexxthiery.github.io/jobs/2024_di...
alexxthiery.github.io/jobs/2024_di...
December 15, 2024 at 2:46 PM
One #postdoc position is still available at the National University of Singapore (NUS) to work on sampling, high-dimensional data-assimilation, and diffusion/flow models. Applications are open until the end of January. Details:
alexxthiery.github.io/jobs/2024_di...
alexxthiery.github.io/jobs/2024_di...
Not completely useless, I think, would be interesting to fine-tune on a library of chess books and games. Somebody must have done that, no?
December 14, 2024 at 12:54 PM
Not completely useless, I think, would be interesting to fine-tune on a library of chess books and games. Somebody must have done that, no?
After watching this beautiful keynote by @arnauddoucet.bsky.social , I *had* to give these Schrodinger bridges a try! Very interesting to be able to "straighten" a basic flow-matching approach. Super cool work by @vdebortoli.bsky.social & co-author!
December 14, 2024 at 11:57 AM
After watching this beautiful keynote by @arnauddoucet.bsky.social , I *had* to give these Schrodinger bridges a try! Very interesting to be able to "straighten" a basic flow-matching approach. Super cool work by @vdebortoli.bsky.social & co-author!
"Mimicking the One-Dimensional Marginal Distributions of Processes Having an Ito Differential" by Gyongy:
link.springer.com/article/10.1...
link.springer.com/article/10.1...
December 13, 2024 at 3:48 PM
"Mimicking the One-Dimensional Marginal Distributions of Processes Having an Ito Differential" by Gyongy:
link.springer.com/article/10.1...
link.springer.com/article/10.1...
Fantastic #neurips keynote by Arnaud Doucet! Really like this slide tracing back many of the modern flow-matching / stochastic interpolants ideas to a 1986 result by probabilist Istvan Gyongy describing how to "Markovianize" a diffusion process (eg. having coefficients depending on all the past)
December 13, 2024 at 3:45 PM
Fantastic #neurips keynote by Arnaud Doucet! Really like this slide tracing back many of the modern flow-matching / stochastic interpolants ideas to a 1986 result by probabilist Istvan Gyongy describing how to "Markovianize" a diffusion process (eg. having coefficients depending on all the past)