anthony goldbloom

anthony goldbloom.png

[san fran]

intro’d to Anthony’s ted 2016 – The jobs we’ll lose to machines — and the ones we won’t

esp resonating: machine: job reducible to frequent high volume tasks.. human: involve tackling novel situations

machine learning – data to prediction… my company has advantage – has looked at tons of data from industry and academia..

industry and academia…? what if we already on a broken feedback loop from those..?

early 90s – started with assessing credit risks.. sorting mail.. now capable of more complex tasks.. ie: 2012 – kaggle built algo to grade high school essay.. last year .. diagnose disease of eye

we have no chance of competing against machines on frequent high volume tasks

machines can’t handle things they haven’t seen many times before.. fundamental limitation of machine learning.. machines need to learn from large volumes of past data..

humans don’t .. we have the ability to connect seemingly disperate threads to solve problems we haven’t seen before..

(percy spencer – microwave from choc bar) – cross polination happens for each of us in small ways 1000s of times per day

machines can’t compete with us when it comes to tackling novel situations.. puts a fundamental limit on the human tasks that machines will automate

future of work – to what extent is that job reducible to frequent high volume tasks and to what extent does it involve tackling novel situations..

on freq high vol.. machines getting smarter and smarter…

tax structuring.. grading essays..

? surely we can use machines for better than that..

but machines not making progress on novel situations..

marketing campaigns… business strategy..

? surely we can use humans for better than that..

let everyday bring you a new challenge.. if it does..  you will stay ahead of the machines..

rev of everyday life ness, ie: a nother way

________

find/follow Anthony:

link twitter

Co-founder and CEO of Kaggle.

https://www.kaggle.com/datasets

initial thoughts on kaggle.. from this page (above).. seems more sustainably emergent if starts from 7 bill plus individual curiosities daily.. rather than going to see what’s already going on.. ie (from page):

Dig in

Explore a dataset with our in-browser analytics tool, Kaggle Kernels. You can also download it in an easy-to-read-format.

Build

Create your data science portfolio. Publish insights and code with Kaggle Kernels and it will be saved to your profile.

Connect

Engage with other data scientists. Share feedback on other Kagglers’ Kernels, or ask a question in a dataset’s forum.

imagining a way (ie: hosting-life-bits via self-talk as data) that curiosity trumps.. what others are already doing.. rather than trying to fit in.. just following your whimsy/fittingness.. the map inside.. your gut

________

same time added:

work

task

job

and perhaps where i’ll focus most of this thinking:

labor page

because of previous interpretive labor page in particular

__________

algorithm ness

 

Advertisements