To help my students comprehend artificial neural networks (ANN), including initialization, backpropagation, and updating, I built an ANN from scratch and developed a custom interactive #Python dashboard using Matplotlib. Now, when I teach these concepts, students engage in

Utterly obsessed with this UCL researcher who won an Ig Nobel for showing that “Blue Zones” where people supposedly live well past 100 at unusual rates are actually just full of clerical errors and people committing pension fraud ucl.ac.uk/ioe/news/2024/…

"[P]enicillin...marked the beginning of a century-long period during which infectious diseases were no longer the leading cause of death. But this…period…seems to be nearing its end…[T]he era of antibiotics [i]s a bubble that is about to burst." science.org/doi/10.1126/sc…

Is Dr. Angela Rasmussen finally acknowledging that the raccoon dogs at Huanan market are almost certainly from Hubei province? Sarbecoviruses with high ACE2 affinity don’t circulate in Hubei but are found in bats 1,000 km south in Yunnan and Laos.

Just a strange juxtaposition by the "Discover more" algorithm

Crits-Christoph et al appears to repeat errors identified a year ago, cites the discredited Pekar et al for multiple spillover and makes overconfident claims based on biased and incomplete early case data: 1. As Ralph Baric says, the outbreak most likely occurred August-Oct 2019

Leave the poor raccoon dogs alone. HSM raccoon dogs were wild-caught in Hubei. None found to have SARS-CoV-2. No link to bats ~1500km away in Yunnan & Laos where the nearest relatives to SARS-CoV-2 are found and Wuhan Institute of Virology sampled SrCoVs. ncbi.nlm.nih.gov/pmc/articles/p…

Louis R Nemzer Forget the new data. Even their own paper said 60/1 not 200/1. Correcting their provable math and coding errors gives 1/5, even before getting into subjective modeling issues!

It seems like the real problem here is that revealing this type of hypocrisy only serves to undermine faith in our leaders, and erodes our morale to follow their dictates. So next time, don't let anyone know 😅 nytimes.com/2024/09/19/nyr…

"Informed participation in societal debates — such as those surrounding technology regulation, climate change, public health, or bioethics — requires a foundational understanding of the relevant scientific principles."  thecrimson.com/article/2024/9…

who.int/news/item/19-0…

Summary of review🧵 by anonymous virologist of Crits-Christoph et al raccoon dog paper: 1. Genotypes of potential hosts were reconstructed for retracing animal geographic origins. None of the genomes correspond to animals known to be present in Yunnan or Laos where the closest

Holtz They also say, off the cuff, it's hard to explain how a virus got to a market with 'so few' human cases in city. Compare a Delta outbreak 2021, Sydney: first cases showed up within a week or two at a mall after just 1 airport driver was infected, who then went shopping ~15km away

And casting out dissidents like Proximal Origin co-author Ian Lipkin after he refused to go along with the unsupported claim COVID-19 arose from Huanan Seafood Market. Lipkin also called their market epicentre analysis "armchair epidemiology based on unverifiable data sets".

"the researchers dubbed it the 'gravito-phononic' effect...a cylinder made from a 4,000-pound aluminum bar and then cool it to its lowest quantum energy state...gravitational waves pass through...distort the cylinder slightly, stretching and squeezing it." bgr.com/science/ground…

Statistical models are odds, polls are percentages. For elections, 55/45 odds in a model is a toss-up, while 55/45 percentage in polls is close to a lock. Thread from 2017 but bears repeating.🤷‍♀️ Size of model fluctuations can NOT be interpreted the same as poll shifts.

The most important Religious debate in History youtu.be/8wVq5aGzSqY?si…