iiinova — where your digital self keeps evolving

Leiv Rønneberg @ltronneberg.bsky.social · 20/09/2026

needs more umlauts for the faux-scandi vibe: Metröpölis-Ådjusted

Reposted by Leiv Rønneberg

Oslo Centre for Biostatistics and Epidemiology @ocbe.bsky.social · 10/09/2026

OCBE's Are Hugo Pripp is organising a Nordic workshop on AI in statistical advising and analysis on 15 September, i.e. next Tuesday. Look forward to an exciting day full of interesting discussion! www.med.uio.no/imb/english/...

med.uio.no

Nordic workshop on AI in statistical advising and analysis - Institute of Basic Medical Sciences

A seminar and workshop on how AI will impact teaching statistics, advising and research methodological in medicine. It is financially supported by incentive funding from Oslo University Hospital.

021

Reposted by Leiv Rønneberg

Jill Walker Rettberg @jilltxt.bsky.social · 26/08/2026

Skal du bevare evnen til å tenke selv må du bli KI-resistent! Her er fem regler for å unngå kognitiv overgivelse! www.khrono.no/fem-regler-f...

khrono.no

Fem regler for KI-resistent forskning

Det er helt nødvendig at vi forskere lærer å gjenkjenne typiske KI-feil både som lesere og som skribenter, skriver KI-forsker Jill Walker Rettberg.

37225

Leiv Rønneberg @ltronneberg.bsky.social · 09/07/2026

Also fun going to ISBA in Nagoya and chatting to someone who already arxived one of our main results two weeks prior (!) arxiv.org/abs/2605.14008 <- excellent work by Torey Hilbert from Ohio State

010

Leiv Rønneberg @ltronneberg.bsky.social · 09/07/2026

Excited about this new pre-print with colleagues in Oslo, providing 'posterior'-type uncertainty for kernel density estimators via the Martingale posterior framework.

131

Reposted by Leiv Rønneberg

Robin Ryder @robinryder.bsky.social · 08/07/2026

I have just resigned from the board of "Statistics and Computing", along with 17 other Associate Editors. This was a difficult decision: this great journal has published many tremendous articles under the leadership of EiC Ajay Jasra, and before him David Hand, Gilles Celeux, and Mark Girolami.

713453

Reposted by Leiv Rønneberg

Eirik Mosveen @mosveen.bsky.social · 26/05/2026

I to og et halvt år har disse to messet om at Oslo står på kanten av det økonomiske stupet. Jeg har ikke tatt det med en klype salt, men en sekk grov havsalt. Les, lik og del (som vi sa i gamle dager).

ao.no

Blakke kjærkerotter som plutselig har fått ræva full av penger. Er ikke det mistenkelig?

Har de snakket sant om pengene våre?

517261

Leiv Rønneberg @ltronneberg.bsky.social · 08/05/2026

Come work with us in Oslo developing Bayesian methodology for personalised treatment decision!

041

Leiv Rønneberg @ltronneberg.bsky.social · 21/03/2026

For all del, men det handler om hvor mye tid vi skal vie til forskjellige aspekter. At maskiner oppnår super-intelligens er etter min mening et smalt tema oppi det hele.

000

Leiv Rønneberg @ltronneberg.bsky.social · 21/03/2026

veldig rar burn av Sterri her i søsterartikkelen om Innsikts engasjement i den offentlige debatten om KI.

110

Leiv Rønneberg @ltronneberg.bsky.social · 21/03/2026

Flott av Inga Strümke i Morgenbladet, setter ord på mye som jeg selv har tenket på det siste året, og som jeg tror mange andre som jobber med ML fra matematikk / statistikk siden tenker.

morgenbladet.no

(+) Inga Strümke: – Spekulativ skit fortrenger viktige diskusjoner

Skrekkscenarioene fra Silicon Valley og tankesmien Langsikt bidrar til PR for KI-bransjen og skygger for viktigere diskusjoner, mener Inga Strümke.

2112

Leiv Rønneberg @ltronneberg.bsky.social · 17/03/2026

I think Mistral's Le Chat should have a Proust mode by default

031

Leiv Rønneberg @ltronneberg.bsky.social · 17/02/2026

From someone who made the switch sociology->maths, there is a really odd «methodology hierarchy» in parts of social science. For some reason quantitative is deemed «harder than» qualitative, therefore better — total nonsense of course. Methods should adapt to research questions. Stick to your guns!

010

Leiv Rønneberg @ltronneberg.bsky.social · 02/11/2025

Yeah, I'm sure there is literature here, I just need to dig a bit! I've been doing some quite crude thresholding so far to decide when things are low-rank 'enough' for the switch. This works sometimes, but seem to also interfere with adaptation of my HMC sampler

020

Leiv Rønneberg @ltronneberg.bsky.social · 02/11/2025

Working on something very much in this vein right now actually. I have a super high-dim problem where due to sparsity the problem becomes low rank during sampling. Wanting to then switch from doing inverses via Cholesky to inverses with Woodbury -- but quite fiddly numerically so far...

120

Leiv Rønneberg @ltronneberg.bsky.social · 31/10/2025

oh thanks! my role was tiny though, the first author (a master's student (!)) is fantastic and did the work

010

Leiv Rønneberg @ltronneberg.bsky.social · 31/10/2025

At a conference this summer a guy referred to papers from the late 90s as «turn of the last century»…

110

Leiv Rønneberg @ltronneberg.bsky.social · 29/10/2025

Been a while since I looked into this but the Hutch++ paper has some optimality results I think as well as some refs to variance reduction techniques arxiv.org/abs/2010.09649 ( from 2021 though, so I'm sure a lot has happened since )

arxiv.org

Hutch++: Optimal Stochastic Trace Estimation

We study the problem of estimating the trace of a matrix $A$ that can only be accessed through matrix-vector multiplication. We introduce a new randomized algorithm, Hutch++, which computes a $(1 \pm ...

130

Leiv Rønneberg @ltronneberg.bsky.social · 27/10/2025

autodiff is cool and all, but have you tried manually computing the gradient to avoid too many variables on the autodiff tape..??? (screenshot of me trying to avoid autodiffing through large Gram matrices)

020

Reposted by Leiv Rønneberg

Henri Pesonen @hpesonen.bsky.social · 01/09/2025

I'm looking for a Doctoral Researcher (PhD student) to work with me on simulation-based inference at Data Science Research Centre, Tampere University Check the link for details and send an application before October 10th. tuni.rekrytointi.com/paikat/?o=A_...

tuni.rekrytointi.com

Doctoral Researcher (simulation-based inference) / Väitöskirjatutkija (simulaatio-pohjainen päättely)

088

Leiv Rønneberg @ltronneberg.bsky.social · 03/07/2025

Playing around plotting some chemical structure embeddings coming from a VAE, and colouring by local intrinsic dimension looks almost like these colour blind test images.

030

Leiv Rønneberg @ltronneberg.bsky.social · 30/05/2025

«– Når det ikkje er nokon risiko for å bli avslørt, og inga form for straff for å lyge, ventar vi frå standard økonomisk teori at alle skulle ha loge, seier Bjorvatn.» Økonom overrasket at folk flest ikke tenker som økonomer

nrk.no

Éin av fire er villig til å lyge for ein tusenlapp

010

Reposted by Leiv Rønneberg

Valeria Vitelli @valeriavitelli.bsky.social · 27/05/2025

JOBS ALERT! PhD position & 3-y postdoc position in statistics within my research group at @ocbe.bsky.social @uio.no Both are linked to a project funded from the Research Council of Norway on Integrative Bayesian clustering for high-dim data in omics Deadline: 13 June 2025 More info & links below!

3119

Leiv Rønneberg @ltronneberg.bsky.social · 02/05/2025

Glad to see this happening! I never understood why UK dads were not marching in the street already when I saw that my UK employer only offered *two weeks* paternity leave. Coming from Norway I had to trade in a minimum 4 months leave for a measly two weeks -- time with my daughter I won't get back.

theguardian.com

British men urged to join ‘Dad strike’ calling for more paternity leave

Exclusive: Fathers planning protest with babies in London on 11 June to highlight UK’s ‘rubbish’ statutory leave, least generous in Europe

032

Leiv Rønneberg @ltronneberg.bsky.social · 21/02/2025

I’d also be interested in this, Martin. How it is structured, what is covered, what the pre-requisites are etc

100

Leiv Rønneberg @ltronneberg.bsky.social · 21/02/2025

I think that could indeed be quite illuminating. Often we approach these topics trying to make them practical without requiring too much mathematical setup. You don’t *need* stochastic process theory to get at the core of GP regression, but you might need it to understand the finer details later on

020

Leiv Rønneberg @ltronneberg.bsky.social · 21/02/2025

Interesting post! I helped run a similar module in Cambridge last year were we tried to cram in too much, covering basics of GPs *and* DPs. If the emphasis is on BNP I think one must mention DPs at some point, though it is a step up in abstraction compared to GPs. «distribution of distributions» etc

230

Leiv Rønneberg @ltronneberg.bsky.social · 03/02/2025

It can be bounded from below and above, by \sqrt{n} and n, respectively. These cases each reflect say an extreme lengthscale, ell=\infty or ell=0, and so it appears almost like a measure of complexity. I'm struggling to come up with an intuitive explanation. Does anyone have any ideas/references?

000

Leiv Rønneberg @ltronneberg.bsky.social · 03/02/2025

Working on a model involving some GP regression, coupled with a horseshoe prior for variable selection. In an attempt at counting the number of effective parameters of my model, the following quantity pops out quite naturally. Is this a well known quantity? 1/2

Here, in the context of GP regression, K refers to the n x n Gram matrix associated with a set of inputs and a covariance function, while lambda_j refers to the eigenvalues of K.

100

Leiv Rønneberg @ltronneberg.bsky.social · 13/01/2025

Maybe I'll add that the call is very general across "computational science", encompassing biology, chemistry, physics etc. alongside maths & stats. Bayesian machine learning is just one of the projects within our department, see the link below www.uio.no/dscience/eng...

uio.no

Mathematics & Statistics - dScience – Centre for Computational and Data Science

Read this story on the University of Oslo's website.

000

Leiv Rønneberg @ltronneberg.bsky.social · 13/01/2025

Second call for the DSTrain Postdoctoral programme @uio.no is up! The university is hiring 18 postdocs across 9 research areas and over 60 loosely defined projects. Come work on Bayesian machine learning with us in beautiful Oslo Deadline April 6th, please share!

uio.no

DSTrain - dScience – Centre for Computational and Data Science

DSTrain is a 5-year postdoctoral programme that will award 36 postdoctoral fellowship positions of 36 months each in two calls over the programme period within the overarching frame of data science. T...

140

Leiv Rønneberg @ltronneberg.bsky.social · 09/01/2025

Are you assuming you are able to compute the derivatives as well, or simply that you know they are non-negative?

110

Leiv Rønneberg @ltronneberg.bsky.social · 26/12/2024

Klassisk økonomfjas. Det er ikke modellene våre eller måleinstrumentene det er noe galt med — folk lider av «fantomfattigdom» borsen.dagbladet.no/nyheter/nord...

121

Leiv Rønneberg @ltronneberg.bsky.social · 17/12/2024

Great department, great group, great city!

021

Reposted by Leiv Rønneberg

Are @litterat.party · 17/12/2024

Tvert i mot, dette er på alle måter et skattemessig smutthull. Det er faktisk nøyaktig hva det er: Noe som teknisk sett ikke er ulovlig, men åpenbart heller ikke i tråd med intensjonen. Intensjonen er neppe at eiendomsbaroner skal få skattefritak.

Engebretsen viser til at Skatteetaten fleire gonger har fått spørsmål om burettslagsmodellen er lovleg, noko dei har stadfesta at han er.

– Dette er ikkje noko skattemessig smotthol på noko som helst måte, sier han.

3405

Leiv Rønneberg @ltronneberg.bsky.social · 10/12/2024

Overbygningen er vel laget av stål..?

010

Leiv Rønneberg @ltronneberg.bsky.social · 05/12/2024

It's quite ironic that the term was first coined in a sociological essay warning *against* meritocracy, rather than lifting it up as an ideal.

en.wikipedia.org

The Rise of the Meritocracy - Wikipedia

120

Leiv Rønneberg @ltronneberg.bsky.social · 05/12/2024

Always fun little tidbits on the wikis of famous mathematicians. Here on Riesz Frigyes' (*of the* Riesz representation theorem) lecturing style

050

Leiv Rønneberg @ltronneberg.bsky.social · 02/12/2024

Yeah, I think for most applications something like a GAM or a GP would be the practical choice compared to a BNN. I'm wondering if the number of layers/parameters needed to get something truly more flexible is just too large to work with

000

Leiv Rønneberg @ltronneberg.bsky.social · 02/12/2024

Hmm, from an initial check __not really__. I initially only fit a single chain, so figured maybe I was exploring a single mode, but when I reran the model with 4 chains I still don't really see much multimodality. Maybe n is so small that the prior dominates?

Posterior samples from weight matrix of the middle layer

000

Leiv Rønneberg @ltronneberg.bsky.social · 01/12/2024

Also, in terms of the sampler complaining, Rhat and effective sample size is awful for most of the parameters (weights), but for the estimated function itself at the input locations it is generally Rhat < 1.02 and n_eff > 300 from 1000 samples. Nice to see these effects myself

100

Leiv Rønneberg @ltronneberg.bsky.social · 01/12/2024

This is a fairly small BNN, "only" ~8k parameters. But given my 100 datapoints, I'm a little surprised that Stan is so efficient. Also a bit disappointed in how smooth the functions look (when sampling from prior), a GP would make quick work of this, even in a full Bayesian treatment.

110