SI485i : NLP Missing Topics and the Future Who cares about NLP? - - PowerPoint PPT Presentation

▶
si485i nlp
SMART_READER_LITE
LIVE PREVIEW

SI485i : NLP Missing Topics and the Future Who cares about NLP? - - PowerPoint PPT Presentation

Nov 27, 2023 131 likes •383 views

SI485i : NLP Missing Topics and the Future Who cares about NLP? NLP has expanded quickly Most top-tier universities now have NLP faculty (Stanford, Cornell, Berkeley, MIT, UPenn, CMU, Hopkins, etc) Commercial NLP hiring: Google,

slide-1
SLIDE 1

SI485i : NLP

Missing Topics and the Future

slide-2
SLIDE 2

Who cares about NLP?

  • NLP has expanded quickly
  • Most top-tier universities now have NLP faculty (Stanford,

Cornell, Berkeley, MIT, UPenn, CMU, Hopkins, etc)

  • Commercial NLP hiring: Google, Microsoft, IBM,

Amazon, LinkedIn, Yahoo

  • Web startups in Silicon Valley are eating up NLP

students

  • Navy, DoD, NSA, NIH: all funding NLP research

2

slide-3
SLIDE 3

What NLP topics did we miss?

  • Speech Recognition

3

slide-4
SLIDE 4

What NLP topics did we miss?

  • Speech Recognition

4

slide-5
SLIDE 5

What NLP topics did we miss?

  • Machine Translation

5

slide-6
SLIDE 6

What NLP topics did we miss?

  • Machine Translation

6

Start at ~6min in. http://www.youtube.com/watch?feature=player_embedded&v=Nu

  • nlQqFCKg
slide-7
SLIDE 7

What NLP topics did we miss?

  • Machine Translation
  • IBM Models (1 through 5)

7

slide-8
SLIDE 8

Machine Translation

  • How to model translations?
  • Words: P( casa | house )
  • Spurious words: P( a | null )
  • Fertility: Pn( 1 | house )
  • English word translates to one Spanish word
  • Distortion: Pd( 5 | 2 )
  • The 2nd English word maps to the 5th Spanish word
slide-9
SLIDE 9

Distortion

  • Encourage translations to follow the diagonal…
  • P( 4 | 4 ) * P( 5 | 5 ) * …
slide-10
SLIDE 10

Learning Translations

  • Huge corpus of “aligned sentences”.
  • Europarl
  • Corpus of European Parliamant proceedings
  • The EU is mandated to translate into all 21 official languages
  • 21 languages, (semi-) aligned to each other
  • P( casa | house ) = (count all casa/house pairs!)
  • Pd( 2 | 5 ) = (count all sentences where 2nd word

went to 5th word)

slide-11
SLIDE 11

Machine Translation Technology

  • Hand-held devices for military
  • Speak english -> recognition -> translation -> generate Urdu
  • Translate web documents
  • Education technology?
  • Doesn’t yet receive much of a focus
slide-12
SLIDE 12

What NLP topics did we miss?

  • Dialogue Systems

12

Do you think Anakin likes me?

I don’t care.

slide-13
SLIDE 13

What NLP topics did we miss?

  • Dialogue Systems
  • Why? Heavy interest in human-robot communication.
  • UAVs require teams of 5+ people for each operating

machine

  • Goal: reduce the number of people
  • Give computer high-level dialogue commands, rather than low-level

system commands

13

slide-14
SLIDE 14

What NLP topics did we miss?

  • Dialogue Systems
  • Dialogue is a fascinating topic. Not only do we need

to understand language, but now discourse cues:

  • Questions require replies
  • Imperatives/Commands
  • Acknowledgments: “ok”
  • Back-channels: “uh huh”, “mm hmm”
  • Belief-Desire-Intention (BDI) Model
  • Beliefs: you maintain a set of facts about the world
  • Desires: things you want to become true in the world
  • Intentions: desires that you are taking action on

14

slide-15
SLIDE 15

What NLP topics did we miss?

  • Unsupervised Learning

15

slide-16
SLIDE 16

What NLP topics did we miss?

  • Unsupervised Learning
  • Most of this semester used data that had human/gold

labels.

  • Bootstrapping was our main counter-example: it is mostly

unsupervised.

  • Many many algorithms being researched to learn

language and knowledge without humans, only using text.

16

slide-17
SLIDE 17

El Fin

  • Secret 1:

17

slide-18
SLIDE 18

El Fin

  • Secret 1:
  • I intentionally made our labs confusing

18

slide-19
SLIDE 19

El Fin

  • Secret 1:
  • I intentionally made our labs confusing

Under-defined tasks with unclear expected results

19

slide-20
SLIDE 20

El Fin

  • Secret 1:
  • I intentionally made our labs confusing

Under-defined tasks with unclear expected results

  • Secret 2:

20

slide-21
SLIDE 21

El Fin

  • Secret 1:
  • I intentionally made our labs confusing

Under-defined tasks with unclear expected results

  • Secret 2:
  • I tried to teach you skills that have nothing to do with NLP

21

slide-22
SLIDE 22

El Fin

  • Secret 1:
  • I intentionally made our labs confusing

Under-defined tasks with unclear expected results

  • Secret 2:
  • I tried to teach you skills that have nothing to do with NLP

Experimentation Error Analysis

22

slide-23
SLIDE 23

El Fin

  • Secret 1:
  • I intentionally made our labs confusing

Under-defined tasks with unclear expected results

  • Secret 2:
  • I tried to teach you skills that have nothing to do with NLP

Experimentation Error Analysis

  • Secret 3:

23

slide-24
SLIDE 24

El Fin

  • Secret 1:
  • I intentionally made our labs confusing

Under-defined tasks with unclear expected results

  • Secret 2:
  • I tried to teach you skills that have nothing to do with NLP

Experimentation Error Analysis

  • Secret 3:
  • I appreciate the hard work you put into the class

24