1. Building a Semantic Parser Overnight

2. Overnight framework

3. Which country has the highest CO2 emissions?
Which had the highest increase since last year?
What fraction is from the five countries with highest GDP?

7. Training data

9. The data problem: The main database is 600 samples (GEO880)
To compare:
Labeled photos: millions
מאיפה הדאטא הקיים מגיע?

10. Not only quantity:
The data can lack critical functionality

11. The process
Domain Seed lexicon Logical forms and canonical utterances Paraphrases Semantic parser

12. The data base:
Triples (e1, p, e2) e1 and e2 are entities (e.g., article1, 2015) p is a property (e.g., publicationDate)

13. Seed lexicon
For every property, a lexical entry of the form <t → s[p]> t is a natural language phrase and s is a syntactic category < “publication date” → RELNP[publicationDate]>

14. Seed lexicon
In addition, L contains two typical entities for each semantic type in the database <alice → NP[alice]>

15. Unary
TYPENP
ENTITYNP
Verb phrases VP ( “has a private bath”)
Binaries:
RELNP functional properties (e.g., “publication date”)
VP/NP transitive verbs (“cites”, “is the president of”)

16. Grammar
<α αn → s[z]> α αn tokens or categories, s is a syntactic category z is the logical form constructed

17. Grammar
<RELNP[r] of NP[x] → NP[R(r).x]> Z: R(publicationDate).article1 C: “publication date of article 1”

18. Crowdsourcing X: “when was article 1 published?”
D = {(x, c, z)} for each (z, c) ∈ GEN(G ∪ L) and x ∈ P(c)

19. Training
log-linear distribution pθ(z, c | x, w)

20. Under the hood

21. Lambda DCS
Entity: singleton set {e} Property: set of pairs (e1, e2)

22. Lambda DCS
binary b and unary u join b.u
𝑒 2 ∈ 𝑢 𝑤 𝑒 1 , 𝑒 2 ∈ 𝑏 𝑤

23. Lambda DCS ¬u
𝑢 1 ∪ 𝑢 2
𝑢 1 ∩ 𝑢 2

24. Lambda DCS
R(b)
(e1, e2) ∈ [b] -> (e2, e1) ∈ [R(b)]

25. Lambda DCS
count(u) sum(u) average(u, b) argmax(u, b)

26. Lambda DCS
λx.u is a set of (e1, e2): e1 ∈ [u[x/e2]]w R(λx.count(R(cites).x)) (e1, e2), where e2 is the number of entities that e1 cites.

27. Seed lexicon for the SOCIAL domain

28. Seed lexicon
article publication date cites won an award

29. Grammar
Assumption 1 (Canonical compositionality): Using a small grammar, all logical forms expressible in natural language can be realized compositionally based on the logical form.

30. Grammar
Functionality-driven Generate superlatives, comparatives, negation, and coordination

31. Grammar

32. Grammar From seed: types, entities, and properties noun phrases (NP)
verbs phrases (VP)
complementizer phrase (CP)
“that cites Building a Semantic Parser Overnight”
“that cites more than three article”

33. Grammar

34. Grammar

35. Grammar

36. Paraphrasing
“meeting whose attendee is alice” ⇒ “meeting with alice” “author of article 1” ⇒ “who wrote article 1” “player whose number of points is 15” ⇒ “player who scored 15 points”

37. Paraphrasing
“article that has the largest publication date ⇒ newest article”. “housing unit whose housing type is apartment ⇒ apartment” “university of student alice whose field of study is music” ⇒ “At which university did Alice study music?”, “Which university did Alice attend?”

38. Sublexical compositionality
“parent of alice whose gender is female ⇒ mother of alice”. “person that is author of paper whose author is X ⇒ co-author of X” “person whose birthdate is birthdate of X ⇒ person born on the same day as X”. “meeting whose start time is 3pm and whose end time is 5pm ⇒ meetings between 3pm and 5pm” “that allows cats and that allows dogs ⇒ that allows pets” “author of article that article whose author is X cites ⇒ who does X cite”.

39. Crowdsourcing in numbers
Each turker paraphrased 4 utterances 28 seconds on average per paraphrase 38,360 responses 26,098 examples remained

40. Paraphrasing noise in the data 17%
noise in the data 17% (“player that has the least number of team ⇒ player with the lowest jersey number”) (“restaurant whose star rating is 3 stars ⇒ hotel which has a 3 star rating”).

41. Model and Learning
numbers, dates, and database entities first

42. Model and Learning
(z, c) ∈ GEN(G ∪ Lx) 𝑝 𝜃 ( z, c | x, w) ∝ exp(φ(c, z, x, w) >θ)

43. Floating parser

44. Floating parser

45. Floating parser

46. Floating parser

47. Model and Learning
Features

48. Model and Learning 𝑥,𝑐,𝑧∈𝐷 𝑙𝑜𝑔 𝑝 θ 𝑧,𝑐 𝑥,𝑤 −𝜆 𝜃 1
𝑥,𝑐,𝑧∈𝐷 𝑙𝑜𝑔 𝑝 θ 𝑧,𝑐 𝑥,𝑤 −𝜆 𝜃 1
AdaGrad (Duchi et al., 2010)

49. Experimental Evaluation