Author: Richard K Collins
Director, The Internet Foundation
Studying formation and optimized collaboration of global communities. Applying the Internet to solve global problems and build sustainable communities. Internet policies, standards and best practices.
Ash Jogalekar @curiouswavefn The paradox of technological innovation – great practical technology develops when people don’t explicitly focus on developing great practical technology. https://pic.twitter.com/9R0vdMIkDS Replying to @curiouswavefn You have it slightly wrong. Innovation goes faster where you do not put too strict bounds on what it will cost, who will use it and how it
Read More »
Machine Learning in Chemistry @ML_Chem Constructing custom thermodynamics using deep learning #machinelearning #compchem https://nature.com/articles/s43588-023-00581-5… Replying to @ML_Chem Machine learning and GPT are all about data preparation, basic statistics, probabilities, regressions, correlations and expected values. They just relabeled it “AI” to make it look cute and new. And the AI groups are too new to know
Read More »
https://twitter.com/RichardKCollin2/status/1741046739237228715 Sensitive three axis gravimeters show that vector sun moon tidal signals are nearly perfectly Newtonian, only requiring three linear regressions (6 numbers) to lock a station to the sun and moon for position and orientation sensing. Many types now. https://twitter.com/RichardKCollin2/status/1693165581329985565 This is one month for the vertical component of the vectors signal.
Read More »
The Science World @scienceworld224 Jupiter and its volcanic moon Io as captured by New Horizons Spacecraft https://pic.twitter.com/eR2CgKrKD2 Replying to @scienceworld224 Thank you for pointing out New Horizons Spacecraft. Reading Wikipedia about it, very satisfying. Vast background knowledge, much on the web, is not connected. GPTs could index “NASA” losslessly. If one wants to do it
Read More »
Big Data Biology Lab @BigDataBiology New preprint: A catalogue of small proteins from the global microbiome! As part of our ongoing efforts to understand small proteins in prokaryotes, we catalogued almost 1 billion sequences https://doi.org/10.1101/2023.12.27.573469 Replying to @BigDataBiology Thank you for this effort. When you put the same text in every row of a file,
Read More »
Mustafa Suleyman @mustafasuleyman Whether it’s commercial, religious, cultural, or military, democratic or authoritarian, every possible motivation can be dramatically enhanced by having cheaper power at your fingertips. These tools will be available to all, a proliferation of not just technology but capability itself https://time.com/6310115/ai-revolution-reshape-the-world/ 9:41 AM · Dec 28, 2023 14.6K Views Who created you?
Read More »
Carlos E. Perez @IntuitMachine 26 Prompting Tips 1 – No need to be polite with LLM so there is no need to add phrases like “please”, “if you don’t mind”, “thank you”, “I would like to”, etc., and get straight to the point. 2 – Integrate the intended audience in the prompt, e.g., the audience
Read More »
@nytimes While @OpenAI and @Microsoft use data that they took without asking, @Google and others routinely “scrape without sharing”. “Knowledge” is shared openly for the good of all, not diverted for the benefit of a few. Individuals sharing with individuals is often FAIR. @huggingface is enabling massive copying. ALL the AIs are not citing their
Read More »
Kirk Borne @KirkDBorne Free 234-page PDF eBook >> Introducing the #Mathematics of Machine Learning: http://bit.ly/3Fe3vMC by @smolix #BigData #DataScience #AI #Algorithms #MachineLearning #DataScientists #DataLiteracy #Statistics #Optimization #LinearAlgebra https://pic.twitter.com/MDnTB6gy5h Replying to @KirkDBorne and @smolix The need on the Internet, is NOT for paper (PDF is paper) that requires human eyes and brains to transcribe, but AI-compatible
Read More »
Pedram Roushan @PedramRoushan Noise destroys long-range correlations on quantum processors and hinders the coherent evolution in the nominally available computational space. Using Random Circuit Sampling we estimate the effective available Hilbert space @GoogleQuantumAI https://arxiv.org/abs/2304.11119 https://pic.twitter.com/TO3yq0IFEn Replying to @PedramRoushan and @GoogleQuantumAI If you record the noise with lossless methods, and high precision, it is likely going
Read More »