Lightnews — Scholar-powered news

Calvin McCarter @calvinmccarter.bsky.social · 8d

i first read that as:

"power washing" but for the feet. idea to fill out later

Calvin McCarter @calvinmccarter.bsky.social · 12d

Slightly off-topic, but one of my biggest pet peeves is when people say about baby sadness, "Don't worry: they won't remember this." In the long run, we are all memoryless!

1

Calvin McCarter @calvinmccarter.bsky.social · Aug 19

What makes tabular data unique (and interesting!) is not merely that it's arranged into rows and columns. New blogpost: calvinmccarter.substack.com/p/the-idiosy...

The idiosyncrasies of tabular data

The things that make tabular data different

calvinmccarter.substack.com

3

Calvin McCarter @calvinmccarter.bsky.social · Aug 1

yea

Calvin McCarter @calvinmccarter.bsky.social · Jul 26

what are the reasons for this from your perspective?

1 1

Calvin McCarter @calvinmccarter.bsky.social · Jun 27

maybe not more expensive to manufacture in an absolute sense. but if chinese internal demand rises, and it no longer needs us exports, then china may no longer see the value in exporting to the us. in which case, the cost of consumer goods *in the us* would rise.

2

Calvin McCarter @calvinmccarter.bsky.social · Jun 27

also, immigrants will also do less of our low-wage / low-status work for us (declining LatAm TFR). meanwhile, AI will (initially) compete with high-wage / high-status workers. so we're still looking at a combustible political situation.

1

Calvin McCarter @calvinmccarter.bsky.social · Jun 27

tbf, while AI is going to start doing more of our work for us, East Asia will start doing less of our work for us (their trade surpluses will decline due to their aging demographics and the end of the USD as reserve currency)

2 1

Calvin McCarter @calvinmccarter.bsky.social · Jun 20

In principle I agree, but federal lands are relatively poorly managed wrt fire prevention and management. Obviously it would be preferable to just fix that, but I'm not sure whether that's realistic.

3

Calvin McCarter @calvinmccarter.bsky.social · Jun 12

Related to your earlier assessment of the probability of the US striking Iran, how would you assess the probability of an Israeli strike? Is it harder to predict, because Israel doesn't need to bring assets into the region?

1

Calvin McCarter @calvinmccarter.bsky.social · May 21

"autoregression for training, diffusion for inference"

4

Calvin McCarter @calvinmccarter.bsky.social · Mar 18

alphaxiv does this for all arxiv papers -- just s/ar/alpha/ in the url -- and i've been told that it's coming soon to {bio,chem,med}rxiv as well.

1 2 4

Calvin McCarter @calvinmccarter.bsky.social · Mar 16

I don't disagree with you exactly, but if an institution's natural defenders are AWOL (or even siding with its enemies) due to a litany of grievances, then it's probably already too late to save that institution.

1 1

Calvin McCarter @calvinmccarter.bsky.social · Mar 16

(not that there's anything wrong with that)

2

Calvin McCarter @calvinmccarter.bsky.social · Mar 11

I just delete my old repo, then give myself a new username...

1

Calvin McCarter @calvinmccarter.bsky.social · Mar 11

Has anyone tried far-UVC in their home? It's now dropped into the ~$300 price range where I'm interested in trying it for myself. substack.com/home/post/p-...

Flipping the switch on far-UVC

We’ve known about far-UVC’s promise for a decade. Why isn't it everywhere?

substack.com

1 2

Calvin McCarter @calvinmccarter.bsky.social · Feb 12

it's definitely a Michigan thing

Calvin McCarter @calvinmccarter.bsky.social · Jan 12

I am slightly cynical about the Clean Label Project, given that it seems to be "pay to play". Also, afaict, it was started by the founder of ByHeart formula, and theirs was like the first thing to get certification. Not that this is necessarily bad -- ByHeart and Kendamil are the best formula IMO.

1

Calvin McCarter @calvinmccarter.bsky.social · Jan 12

the website has a list: cleanlabelproject.org/product-cate... which is specific and helpful, though oddly it slightly differs from the brands in the report:

1

Calvin McCarter @calvinmccarter.bsky.social · Jan 12

Here's a link to the report: cleanlabelproject.org/wp-content/u... (TLDR heuristics: whey is better than plant-based, non-organic is better than organic, unflavored is better than chocolate-flavored)

1

Calvin McCarter @calvinmccarter.bsky.social · Jan 8

Ultra exciting! And it's gratifying to see that this method uses the kernel density integral preprocessing method that I published in @tmlr-pub.bsky.social (2023). (One takeaway: even if your ML research focus isn't deep learning, pursue directions that complement rather than compete with it.)

Samuel Müller @sammuller.bsky.social · Jan 8

This might be the first time after 10 years that boosted trees are not the best default choice when working with data in tables.
Instead a pre-trained neural network is, the new TabPFN, as we just published in Nature 🎉

2

Calvin McCarter @calvinmccarter.bsky.social · Dec 23

i cooked him

3

Calvin McCarter @calvinmccarter.bsky.social · Dec 20

There are ways around this, eg arxiv.org/abs/1902.04094, but I think uniform rate MLMs are the cleaner approach. Plus, as shown in Eq 5 of the Factorization Curse paper, it's what you get after marginalizing over all XLNet-style autoregressive permutations.

BERT has a Mouth, and It Must Speak: BERT as a Markov Random Field Language Model

We show that BERT (Devlin et al., 2018) is a Markov random field language model. This formulation gives way to a natural procedure to sample sentences from BERT. We generate from BERT and find that it...

arxiv.org

1

Calvin McCarter @calvinmccarter.bsky.social · Dec 20

This creates a train-inference gap if one is training at any fixed masking rate. Of course, with MLMs people don't even try that, and instead use pseudo-likelihood as an approximation for likelihood. Besides being approximate, the problem with this is that this takes L separate forward-passes.

1 1

Calvin McCarter @calvinmccarter.bsky.social · Dec 20

When one evaluates log-likelihood of a sequence of length L via the chain rule of probability, the first term has missingness fraction of 1, the second has missingness of (L-1)/L, etc. So the inference-time masking rate is ~ Uniform[0, 1].

2 1 1