The “Big Picture” Works for language too:

The “Big Picture”
Works for language too:
Shll cmpr th t smmr’s d?
Th rt mr lvl nd mr tmprt.
Rgh wnds d shk th drlng bds f M,
nd smmr's ls hth ll t shrt dt.
Smtm t ht th f hvn shns,
nd ftn s hs gld cmplxn dmm'd;
nd vr fr frm fr smtm dclns,
B chnc r ntr's chngng crs ntrmm'd;
Bt th trnl smmr shll nt fd
Nr ls pssssn f tht fr th w'st;
Nr shll Dth brg th wndr'st n hs shd,
Whn n trnl lns t tm th grw'st:
S lng s mn cn brth r s cn s,
S lng lvs ths, nd ths gvs lf t th.
Shall I compar thee t a smmer's dy?
Thou rt mor lovly and mre temprate.
Rogh winds do shke th darlng buds of May,
And smmer's leas hath all too shrt a date.
Somtimes too hot th eye of heavn shines,
And oftn is hs gold complxion dimm'd;
And evry fair frm fair somtime declines,
By chnce or natur's changng cours untrimm'd;
But thy etrnal summer shll not fade
Nor lose pssession of tht fair thou ow'st;
Nr shll Death brag thou wander'st in hs shade,
Whn in eternl lines to time thou grow'st:
So long as men cn breathe or eys cn see,
So lng lives this, nd this givs lif to thee.
Shall I compare thee to a summer's day?
Thou art more lovely and more temperate.
Rough winds do shake the darling buds of May,
And summer's lease hath all too short a date.
Sometime too hot the eye of heaven shines,
And often is his gold complexion dimm'd;
And every fair from fair sometime declines,
By chance or nature's changing course untrimm'd;
But thy eternal summer shall not fade
Nor lose possession of that fair thou ow'st;
Nor shall Death brag thou wander'st in his shade,
When in eternal lines to time thou grow'st:
So long as men can breathe or eyes can see,
So long lives this, and this gives life to thee.
WILLIAM SHAKESPEARE (1564-1616)
What’s the point?
Your brain is really good at language and vision, even
under adverse conditions.
You weren’t born knowing this information, or how to
read: you learned those things.
How might (a computer simulation of) your brain be
good at vision, language, and learning?
So wait 15 years…
Hidden units solve XOR, and any other function
Back-propagation takes care of learning
Lots of successful applications: NETTalk, Speech
Recognition, Past Tense model, classifier systems
NN’s aren’t like brains in detail, but are much more like them
than are grammars, programs, etc.