I had an interesting conversation with a colleague this week. He has built a model that predicts physiological stability for inpatients on intravenous antibiotics using a recent history of vital signs. The purpose is to identify patients who could be switched to oral antibiotics and therefore discharged sooner.
I had also considered that the underlying physiological stability, or its inverse, was a common component of many acute hospital prediction problems. But I wonder if we could take this further when considering how we build features under the ML-Ops paradigm.
We are entering an era where someone might use a large language model to generate a document out of a bulleted list, and send it to a person who will use a large language model to condense that document into a bulleted list. Can anyone seriously argue that this is an improvement?
Training may want the large data center, but we’ve long since hit the point of diminishing returns. There is effectively no more text to train on, as even the LLM systems of a few years ago were trained on almost all the coherent text in existence
go grab a spreadsheet and figure out how to get the best result on a boring economics problem with a boring algorithm; Don’t worry so much about making a painting or movie with GANs, we’re already really good at doing that and enjoy doing it.
Store git credentials for a day from the command line.
As of Aug, 2021, a ‘personal access token’ must replace your password.
Link list of programming tips
A crowd-sourced list of recommendations for learning good programming style from the lab group.
Software Design in Python Pragmatic Programmer Clean Code Philosophy of Software Design Code Craft The Missing README Beyond the Basic Stuff Refactoring code All the little things Nothing is something
I just discovered that iA Writer publishes to micro.blog. How cool! Am going to see if I can use this to improve my writing habits as per Andrew Gelman's recent comment:
Eventually everything makes its way from conversation to blog to publication. A great philosophy.
What’s the purpose of mathematical modeling? « Statistical Modeling, Causal Inference, and Social Science
Scientists — not just in epidemiology, but in physics, ecology, climatology, economics and every other field — don’t build models as oracles of the future. For them, a model “is just a way of understanding a particular process or a particular question we’re interested in,” Kucharski said, “and working through the logical implications of our assumptions.
Not just a cool scientific advance, but a very cool explanation of a new technique. For that matter, a lovely explanation of matrix multiplication for the non-mathematician.
New Algorithm Breaks Speed Limit for Solving Linear Equations By harnessing randomness, a new algorithm achieves a fundamentally novel — and faster — way of performing one of the most basic computations in math and computer science.
Quanta Magazine quantamagazine.orgquantamagazine.org https://www.quantamagazine.org/new-algorithm-breaks-speed-limit-for-solving-linear-equations-20210308/
