AlphaFold: a solution to a 50-year-old grand challenge in biology

Re:AlphaFold: a solution to a 50-year-old grand challenge in... by whimsicalism on Monday November 30, 2020 @11:04AM (#25255153)

ML is a super overloaded term.

There are definitely cases where machine learned statistical solutions do not perform as well as the systems tuned by the experts, but if you can define the task well and get the data for a deep solution, usually those will overtake.

[ Parent | on HN ]

Re:AlphaFold: a solution to a 50-year-old grand challenge in... by penagwin on Monday November 30, 2020 @12:29PM (#25256332)

This. I believe technically just linear regression could be considered "machine learning".

[ Parent | on HN ]

Re:AlphaFold: a solution to a 50-year-old grand challenge in... by misnome on Monday November 30, 2020 @12:56PM (#25256698)

I've seen people at bio conferences actively calling linear regression machine learning.

[ Parent | on HN ]

Re:AlphaFold: a solution to a 50-year-old grand challenge in... by diab0lic on Monday November 30, 2020 @01:51PM (#25257404)

This is likely because linear regression meets most widely accepted definitions of machine learning. [0][1] It is simple and very effective when learning in linear space.

[0] https://en.wikipedia.org/wiki/Machine_learning

[1] https://www.cs.cmu.edu/~tom/mlbook.html

[ Parent | on HN ]

Re:AlphaFold: a solution to a 50-year-old grand challenge in... by b3kart on Monday November 30, 2020 @06:22PM (#25260350)

Sorry, I don’t get it. Are you saying fitting a linear regression model to data and making predictions somehow isn’t machine learning? I am confused.

[ Parent | on HN ]