«   2018/09   »
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30            
Archives
Today
12
Total
44,940
관리 메뉴

breaktime

The current state of Machine Intelligence 3.0 요약 본문

엔지니어링/머신러닝

The current state of Machine Intelligence 3.0 요약

블루제리 2017.10.16 12:15

원문: https://www.oreilly.com/ideas/the-current-state-of-machine-intelligence-3-0


"One stop shop" of the machine intelligence stack is coming into view.  

This just scratches the surface of all of the activity out there.


Ready player world

The gaming world offers a perfect place to start machine intelligence work 

(e.g., constrained environments, explicit rewards, easy-to-compare results) 

We're excited to see what new creative applications come in the next year.


Why even bot-her?

Two componets of bot technology: the conversational interface and "agent" 

(learning from data and transacting on a user's behalf).

There will not be one agent to rule them all, even if there is one interface to rule them all.

Machine intelligence will let us all work like CEOs. 

(https://hbr.org/2016/06/machine-intelligence-will-let-us-all-work-like-ceos)



On to 1111100001

Succesful use of machine intelligence is surprisingly binary like flipping a stubborn light switch.  

Organization will see everything through the lens of potential of machine intelligence.

Unlike traditional software, machine intelligence gives only probabilistic outputs. 

The biggest predictor of whether a company will successfully adopt machine intelligence is whether they have a C-Suite executive with an advanced math degree.

The real danger of machine intelligence is that executives will make bad decisions about what machine intelligence capabilities to build.


Peter Pan's never-never land

Acquisitions by big companies (Apple, Samsung, GM, etc.) and the absence of a market in the machine intelligence make difficult for a machine intelligence startup to "grow up" - hence the Peter Pans.

Companies successfully entering a long-term trajectory can package their technology as a new problem-specific application for enterprise or simply transform an industry themselves as a new entrant (love this)


Inspirational machine intelligence

Machine intelligence can solve much more intersting problems than traditional software for good.


Looking forward

Once we understand machine intelligence deeply, we might look back on the era of traditional software and think it was just a prologue to what's happening now.


'엔지니어링 > 머신러닝' 카테고리의 다른 글

The current state of Machine Intelligence 3.0 요약  (0) 2017.10.16
0 Comments
댓글쓰기 폼