Posts Tagged ‘00x27ips’

Hillary Owes Americans the Truth About the Cause of Her Concussion. Trust Is Earned. | RedState

Sunday, February 21st, 2016

Truth About the Cause of [Clinton’s] Concussion Wild post on a #conspiracy theory that’ll make a great movie plot


A Deadly Deployment, a Navy SEAL’s Despair – The New York Times

Wired 12.06: Cracking the Code to Romance

Sunday, June 15th, 2014

Interesting mention of bioinformatics

NHS patients set for genetic check-up

Monday, July 29th, 2013

Two foci for Genomics England are cancer and rare, inherited diseases – not neuro, cardio, &c.

NHS patients set for genetic check-up

Saturday, July 27th, 2013

It’s Not You, It’s Your Books – NY Times

Monday, July 22nd, 2013

A Short Proust and a Long Bellow, Harvard Magazine

Sunday, July 21st, 2013

Biographer and editor James Atlas tells stories of real lives by Craig Lambert
September-October 2000

NHGRI to Unveil Funding for Tools to Interpret Non-coding Genomic Regions

Saturday, May 25th, 2013

Ken Auletta: At Business Insider, Henry Blodget’s Second Act : The New Yorker

Saturday, April 13th, 2013

class of ’88 at Yale & onto millions on Wall St. before an
investigation and now a 2nd act

Thoughts on “A few useful things to know about machine learning”

Thursday, February 14th, 2013

Some thoughts on a good paper giving intuition on machine learning approaches

In particular, the paper gives good intuition about:

– overfitting (e.g. how it’s related to multiple testing & bias v variance)
– the curse of dimensionality (in high-D all neighbors look the same)
– the non-practicality of theoretical guarantees
– how different frontiers can give the same prediction
– ensembles (which reduce variance greatly without increasing bias that much)
– ensembles vs Bayesian model averaging (which essentially select the best model)

A few useful things to know about machine learning

Saturday, February 9th, 2013