Tootfinder

SUCCESS!! Just got confirmation from the Spanish Consultate in Toronto of an appointment in Vancouver to complete the kids’ passport applications (fingerprints) on April 12.
It's happening! They're gonna have Spanish/EU passports!!!
(And we successfully avoided the cost/CO2/hassle of flying to Toronto!)
#Spain

Very excited to see this collaboration between 2 of the most interesting research & scholarly communication infrastructure initiatives out there. This stellar cast of scholars & infrastructure builders will undoubtedly be pushing the limits in fascinating ways. This is a project worth following. Bravo!
From: @…

COPIM (@copim@hcommons.social)
The second pilot project is: Database as Book and Lively Community Archive: https://copim.pubpub.org/pub/database-book-launch which will build a “database as book” to support an emergent collective of critical STS scholars in Kenya in collaboration with the Research Data Share Collective, African Minds, and Platform for Experimental Collaborative Ethnography (PECE).

“Let ’em eat ’flakes”
https://www.today.com/food/groceries/kellogg-ceo-cereal-for-dinner-save-money-backlash-rcna140540

Kellogg CEO Says People Should Eat ‘Cereal for Dinner’ to Save Money
WK Kellogg CEO Gary Pilnick’s cost-saving suggestion of eating cereal for dinner sparks backlash from consumers who are feeling the strain of grocery prices.

I'd be interested in a general upwelling of feeling to abolish ICO and rely on the Civil Court with zero cost suits for individuals to bring TORT cases against offenders, which would also allow joint litigants.
#RegulatorsImpedeRegulation

Neural Architecture Search for Sentence Classification with BERT
Philip Kenneweg, Sarah Schr\"oder, Barbara Hammer
https://arxiv.org/abs/2403.18547 ht…

Neural Architecture Search for Sentence Classification with BERT
Pre training of language models on large text corpora is common practice in Natural Language Processing. Following, fine tuning of these models is performed to achieve the best results on a variety of tasks. In this paper we question the common practice of only adding a single output layer as a classification head on top of the network. We perform an AutoML search to find architectures that outperform the current single layer at only a small compute cost. We validate our classification architec…

I finally caught up with >1 week of notifications in Mastodon! Sorry for the delay.
Can I help you with anything?

This https://arxiv.org/abs/2306.13185 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_sta…

An Agnostic View on the Cost of Overfitting in (Kernel) Ridge Regression
We study the cost of overfitting in noisy kernel ridge regression (KRR), which we define as the ratio between the test error of the interpolating ridgeless model and the test error of the optimally-tuned model. We take an "agnostic" view in the following sense: we consider the cost as a function of sample size for any target function, even if the sample size is not large enough for consistency or the target is outside the RKHS. We analyze the cost of overfitting under a Gaussian universality an…

Just posted the following on #nextdoor Very curious about the responses if there are any.
"Why doesn't Highway Unit 1 enforce the 50mph speed limit on the West Side Highway? There are so many collisions there caused by drivers obviously going way more than 50mph. So many sirens. So much cost. I would assume all that expense raises car insurance rates. People are always complaining …

An apt upgrade on Elementary has caused an explosion of language packs and splattering of dpkg talking to itself. It ended by trying to install nginx and failing which I don't actually care about.
We shall see what we shall see.

This https://arxiv.org/abs/2306.13185 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_sta…

An Agnostic View on the Cost of Overfitting in (Kernel) Ridge Regression
We study the cost of overfitting in noisy kernel ridge regression (KRR), which we define as the ratio between the test error of the interpolating ridgeless model and the test error of the optimally-tuned model. We take an "agnostic" view in the following sense: we consider the cost as a function of sample size for any target function, even if the sample size is not large enough for consistency or the target is outside the RKHS. We analyze the cost of overfitting under a Gaussian universality an…