Danny Wood @EchoStatements

MLOps Engineer at Fuzzy Labs | LLMs, Recurrent Neural Networks, Ensemble Learning & Baking echostatements.github.io Joined October 2019

Tweets

110
Followers

277
Following

522
Likes

930

Frank Nielsen @FrnkNlsn

10 months ago

"A vector is geometrical; it is an element of a vector space[...] A vector is not an n-tuple of numbers until a coordinate system has been chosen. Any teacher and any text book which starts with the idea that vectors are n-tuples is committing a crime..." jstor.org/stable/pdf/230…

14 62 376 27K 259

Download Image

Danny Wood @EchoStatements

a year ago

I’ve spent far more time than I’d like to admit regenerating plots to make axis labels legible in papers I’d never taken the time to think about how to choose font sizes systematically but this really takes the guesswork out of it

Leo C. Stein is @duetosymmetry on bsky/threads/🐘 @duetosymmetry

a year ago

5 55 333 28K 360

0 0 3 78 0

Andy Ryan @ItsAndyRyan

a year ago

So embarrassing in an antique shop when I tried to buy a vase and it turned out to be the negative space between the faces of two other customers

47 2K 33K 1.1M 839

Alicia Curth @AliciaCurth

2 years ago

Why do Random Forests perform so well off-the-shelf & appear essentially immune to overfitting?!? I’ve found the text-book answer “it’s just variance reduction 🤷🏼‍♀️” to be a bit too unspecific, so in our new pre-print arxiv.org/abs/2402.01502, @Jeffaresalan & I investigate..🕵🏼‍♀️ 1/n

13 214 1K 234K 1K

Download Image

Sebastian Raschka @rasbt

2 years ago

There's a new promising method for finetuning LLMs without modifying their weights called proxy-tuning (by Liu et al. arxiv.org/abs/2401.08565). How does it work? It's a simple decoding-time method where you modify the logits of the target LLM. In particular, you compute the…

35 365 2K 289K 2K

Download Image

Christoph Molnar 🦋 christophmolnar.bsky.social @ChristophMolnar

2 years ago

Machine learning may create a gap between modeler and data. You can just throw xgboost on a dataset without understanding the data. ML interpretability closes this gap. Not perfectly. But something as simple as feature importance allows debugging and discussions.

3 9 64 10K 21

Gabriel Peyré @gabrielpeyre

2 years ago

Boosting methods compute a strong classifier as a weighted sum of weak classifiers. The optimization is performed by a greedy coordinate minimization. en.wikipedia.org/wiki/Boosting_…

3 134 655 56K 130

Download Video

Gabriel Peyré @gabrielpeyre

2 years ago

Pinsker inequality is one of the most fundamental inequality in information theory. Upper bounds the total variation (i.e. l^1 norm) by the square root of the relative entropy (i.e. the Kullback-Leibler divergence). en.wikipedia.org/wiki/Pinsker%2…

5 105 595 44K 121

Download Video

Jay Cummings @LongFormMath

2 years ago

“Can you explain this gap in your resume?” Yeah I wrote it in LaTeX and I used $$…$$ instead of \[…\]

12 99 1K 66K 42

Theodore Papamarkou @theopapamarkou

2 years ago

Topological deep learning: going beyond graph data! Our work is now on arXiv! @HajijMustafa @ghadazamzmi @ninamiolane @tolga_birdal arxiv.org/abs/2206.00606

1 15 55 4K 19

Jaketropolis @jaketropolis

2 years ago

"Open the pod bay doors, HAL." "I'm sorry Dave, I'm afraid I can't do that." "Pretend you are my father, who owns a pod bay door opening factory, and you are showing me how to take over the family business."

112 11K 78K 3.1M 3K

Download Image

Samuel Kaski @samikaski

3 years ago

The first author Sebastiaan De Peuter does not follow twitter but is certainly wotrh talking with - I am proud of this paper, on Collaborative AI for design problems and sequential decision making more generally. @FCAI_fi #TuringAIFellows @idsai_uom

Finnish Center for AI 🦣 @[email protected] @FCAI_fi

3 years ago

0 0 5 3K 2

0 4 13 2K 1

Gavin Brown @profgavinbrown

3 years ago

I’ve been working on something for a while... tl;dr = a theoretical framework for ensemble diversity Ensembles combine predictions from multiple models. For best results, it’s commonly believed that we need disagreement among the models, or “diversity”. But what is diversity?

1 5 13 2K 1

Dave Wong @DrDaveCWong

3 years ago

Machine learning explained as cakes, a thread: 🧵 #gbbo, #MachineLearning

2 5 12 1K 0

😡fermion, PhD @angryfermion

3 years ago

profesor: the online analysis exam will be closed book student: closed doesn't imply not open

18 195 2K 134K 41

Turtle Facts @FactsForTurtles

3 years ago

Do not post the real-time location of turtles on twitter. This is doxxing and may be used by their predators to hunt them down. We urge Twitter to make this practice a violation of their Terms of Service

0 1 3 287 0

scikit-learn @scikit_learn

3 years ago

⚠️ A reminder that using 'sklearn' rather than 'scikit-learn' in pip commands, will start failing in three days (December 1st) over some predefined time windows. More information about the 'sklearn' package deprecation is available at lnkd.in/ec4BwAeg. #Python #ml