building adaptable and scalable natural language
play

Building Adaptable and Scalable Natural Language Generation Systems - PowerPoint PPT Presentation

Mar 10, 2024 •39 likes •1.78k views

Building Adaptable and Scalable Natural Language Generation Systems Yannis Konstas Natural Language Generation is everywhere (Machine Translation)

  1. Sequence to sequence model input output Encoder Decoder a I know inhabit Attention planet The knew inhabited man A planet was … … … … … know ARG0 I ARG1 ( planet ARG1-of inhabit Y w i | w <i , h ( s ) � � w = argmax ˆ p <s> w i I know the planet of

  2. Linearization Graph —> Depth First Search hold ARG0 ARG1 time location person city date-entity meet ARG0-of ARG0 year month name have-role person ARG1 ARG2 2002 1 “New York” ARG1-of ARG2-of official country name expert group “United States” US officials held an expert group meeting in January 2002 in New York .

  3. Linearization Graph —> Depth First Search hold hold ARG0 ARG1 time location :ARG0 (person :ARG0-of (have-role person city :ARG1 United_States date-entity meet :ARG2 official) ARG0-of ) ARG0 year month name :ARG1 (meet have-role :ARG0 (person person ARG1 ARG2 :ARG1-of expert 2002 1 “New York” ARG1-of ARG2-of :ARG2-of group) official country ) name :time (date-entity 2002 1) expert group :location New_York “United States” US officials held an expert group meeting in January 2002 in New York .

  4. Encoding Linearize —> RNN encoding hold :ARG0 (person :ARG0-of (have-role :ARG1 United_States :ARG2 official) ) :ARG1 (meet :ARG0 (person :ARG1-of expert :ARG2-of group) ) :time (date-entity 2002 1) :location New_York

  5. Encoding Linearize —> RNN encoding hold :ARG0 (person :ARG0-of (have-role :ARG1 United_States :ARG2 official) ) :ARG1 (meet :ARG0 (person :ARG1-of expert :ARG2-of group) ) :time (date-entity 2002 1) :location New_York

  6. Encoding Linearize —> RNN encoding Token embeddings - hold :ARG0 (person :ARG0-of (have-role :ARG1 United_States :ARG2 official) ) :ARG1 (meet :ARG0 (person :ARG1-of expert :ARG2-of group) ) hold ARG0 ( person ARG0-of :time (date-entity 2002 1) :location New_York

  7. Encoding Linearize —> RNN encoding Token embeddings - Recurrent Neural Network (RNN) - hold :ARG0 (person :ARG0-of (have-role :ARG1 United_States h 1(s) h 2(s) h 3(s) h 4(s) h 5(s) :ARG2 official) ) :ARG1 (meet :ARG0 (person :ARG1-of expert :ARG2-of group) ) hold ARG0 ( person ARG0-of :time (date-entity 2002 1) :location New_York

  8. Encoding Linearize —> RNN encoding Token embeddings - Recurrent Neural Network (RNN) - Bi-directional RNN - hold :ARG0 (person :ARG0-of (have-role :ARG1 United_States h 1(s) h 2(s) h 3(s) h 4(s) h 5(s) :ARG2 official) ) :ARG1 (meet :ARG0 (person :ARG1-of expert :ARG2-of group) ) hold ARG0 ( person ARG0-of :time (date-entity 2002 1) :location New_York

  9. Encoding Linearize —> RNN encoding Token embeddings - Recurrent Neural Network (RNN) - Bi-directional RNN - [ ] [ ] [ ] [ ] [ ] h 1(s) h 2(s) h 3(s) h 4(s) h 5(s) hold :ARG0 (person :ARG0-of (have-role :ARG1 United_States h 1(s) h 2(s) h 3(s) h 4(s) h 5(s) :ARG2 official) ) :ARG1 (meet :ARG0 (person :ARG1-of expert :ARG2-of group) ) hold ARG0 ( person ARG0-of :time (date-entity 2002 1) :location New_York

  10. Encoding Linearize —> RNN encoding Token embeddings - Recurrent Neural Network (RNN) - Bi-directional RNN - [ ] [ ] [ ] [ ] [ ] h 1(s) h 2(s) h 3(s) h 4(s) h 5(s) hold :ARG0 (person :ARG0-of (have-role :ARG1 United_States h 1(s) h 2(s) h 3(s) h 4(s) h 5(s) :ARG2 official) ) :ARG1 (meet :ARG0 (person :ARG1-of expert :ARG2-of group) ) hold ARG0 ( person ARG0-of :time (date-entity 2002 1) :location New_York

  11. Decoding RNN Encoding —> RNN Decoding (Beam search) h 1 h N(s)

  12. Decoding RNN Encoding —> RNN Decoding (Beam search) init h ( s ) - h 1 h N(s) ∅

  13. Decoding RNN Encoding —> RNN Decoding (Beam search) init h ( s ) Holding - Held softmax - US … h 1 h N(s) ∅

  14. Decoding RNN Encoding —> RNN Decoding (Beam search) init h ( s ) Holding a - Held the softmax - US meeting w i | w <i , h ( s ) � � … … - p h 1 h 2 h N(s) ∅ w 11: Holding Helds w 12: Hold w 13: US w 14: …

  15. Decoding RNN Encoding —> RNN Decoding (Beam search) US init h ( s ) Holding a - person Held the softmax expert - US meeting … w i | w <i , h ( s ) � � … … - p … h 1 h 2 h 3 h N(s) ∅ w 11: Holding Hold a w 21: Helds w 12: w 22: Hold the Hold w 13: Held a w 23: US w 14: Held the w 24: … …

  16. Decoding RNN Encoding —> RNN Decoding (Beam search) US init h ( s ) Holding a meeting - person Held the meetings softmax expert - US meeting meet … w i | w <i , h ( s ) � � … … … - p … h 1 h 2 h 3 h k h N(s) ∅ officials held w k1: The US w 11: Holding Hold a w 21: officials held a US w k2: Helds w 12: w 22: Hold the Hold w 13: Held a w 23: US officials hold the w k3: US w 14: Held the w 24: officials will hold a w k4: US … … …

  17. Attention a the meeting … h 2 h 3 w 2 : held

  18. Attention w 2 : held a the meeting [ ] [ ] [ ] [ ] [ ] … h 3 h 1(s) hold h 2(s) ARG0 [ ] c 3 h 3(s) ( person h 4(s) h 5(s) ARG0-of

  19. Attention w 2 : held a the meeting [ ] [ ] [ ] [ ] [ ] … h 3 h 1(s) hold h 2(s) ARG0 [ ] c 3 h 3(s) ( person h 4(s) h 5(s) ARG0-of h ( s ) , h i � � �� a i = soft max f i a ij h ( s ) X c i = j i

  20. Attention w 2 : held a the meeting [ ] [ ] [ ] [ ] [ ] … h 3 h 1(s) hold h 2(s) ARG0 [ ] c 3 h 3(s) ( person h 4(s) hold ARG0 ( person role US official ) ARG1 ( meet expert group ) US h 5(s) ARG0-of officials held an expert h ( s ) , h i � � �� a i = soft max f i group meeting a ij h ( s ) X c i = in j January i 2002

  21. Attention w 2 : held a the meeting [ ] [ ] [ ] [ ] [ ] … h 3 h 1(s) hold h 2(s) ARG0 [ ] c 3 h 3(s) ( person h 4(s) hold ARG0 ( person role US official ) ARG1 ( meet expert group ) US h 5(s) ARG0-of officials held an expert h ( s ) , h i � � �� a i = soft max f i group meeting a ij h ( s ) X c i = in j January i 2002

  22. Attention w 2 : held a the meeting [ ] [ ] [ ] [ ] [ ] … h 3 h 1(s) hold h 2(s) ARG0 [ ] c 3 h 3(s) ( person h 4(s) hold ARG0 ( person role US official ) ARG1 ( meet expert group ) US h 5(s) ARG0-of officials held an expert h ( s ) , h i � � �� a i = soft max f i group meeting a ij h ( s ) X c i = in j January i 2002

  23. Attention w 2 : held a the meeting [ ] [ ] [ ] [ ] [ ] … h 3 h 1(s) hold h 2(s) ARG0 [ ] c 3 h 3(s) ( person h 4(s) hold ARG0 ( person role US official ) ARG1 ( meet expert group ) US h 5(s) ARG0-of officials held an expert h ( s ) , h i � � �� a i = soft max f i group meeting a ij h ( s ) X c i = in j January i 2002

  24. Pre-processing Linearization —> Anonymization hold hold ARG0 ARG1 time location :ARG0 (person :ARG0-of (have-role person city :ARG1 loc_0 date-entity meet :ARG2 official) ARG0-of ) ARG0 year month name :ARG1 (meet have-role :ARG0 (person person ARG1 ARG2 :ARG1-of expert 2002 1 “New York” ARG1-of ARG2-of :ARG2-of group) official country ) name :time (date-entity year_0 month_0) expert group :location loc_1 “United States” US officials held an expert group meeting in January 2002 in New York .

  25. Pre-processing Linearization —> Anonymization hold hold ARG0 ARG1 time location :ARG0 (person :ARG0-of (have-role person city :ARG1 loc_0 date-entity meet :ARG2 official) ARG0-of ) ARG0 year month name :ARG1 (meet have-role :ARG0 (person person ARG1 ARG2 :ARG1-of expert 2002 1 “New York” ARG1-of ARG2-of :ARG2-of group) country official country ) name :time (date-entity year_0 month_0) expert group :location loc_1 “United “United States” States” US officials held an expert group meeting in January 2002 in New York .

  26. Pre-processing Linearization —> Anonymization hold hold ARG0 ARG1 time location :ARG0 (person :ARG0-of (have-role person city :ARG1 loc_0 date-entity meet :ARG2 official) ARG0-of ) ARG0 year month name :ARG1 (meet have-role :ARG0 (person person ARG1 ARG2 :ARG1-of expert 2002 2002 1 1 “New York” ARG1-of ARG2-of :ARG2-of group) country official country ) name :time (date-entity year_0 month_0) expert group :location loc_1 “United “United States” States” US officials held an expert group meeting in January 2002 in New York .

  27. Pre-processing Linearization —> Anonymization hold hold ARG0 ARG1 time location :ARG0 (person :ARG0-of (have-role person city city :ARG1 loc_0 date-entity meet :ARG2 official) ARG0-of ) ARG0 year month name :ARG1 (meet have-role :ARG0 (person person ARG1 ARG2 :ARG1-of expert 2002 2002 1 1 “New York” “New York” ARG1-of ARG2-of :ARG2-of group) country official country ) name :time (date-entity year_0 month_0) expert group :location loc_1 “United “United States” States” US officials held an expert group meeting in January 2002 in New York .

  28. Pre-processing Linearization —> Anonymization hold hold ARG0 ARG1 time location :ARG0 (person :ARG0-of (have-role person city city :ARG1 loc_0 date-entity meet :ARG2 official) ARG0-of ) ARG0 year month name :ARG1 (meet have-role :ARG0 (person person ARG1 ARG2 :ARG1-of expert 2002 2002 1 1 “New York” “New York” ARG1-of ARG2-of :ARG2-of group) country official country ) name :time (date-entity year_0 month_0) expert group :location loc_1 “United “United States” States” US officials held an expert group meeting in January 2002 in New York . loc_0 officials held an expert group meeting in month_0 year_0 in loc_1 .

  29. Experimental Setup AMR LDC2015E86 (SemEval-2016 Task 8) ‣ Hand annotated MR graphs: newswire, forums ‣ ~16k training / 1k development / 1k test pairs Train ‣ Optimize cross-entropy loss Evaluation ‣ BLEU n-gram precision (Papineni et al., ACL 2002)

  30. First Attempt TreeToStr : Flanigan et al, NAACL 2016 TSP : Song et al, EMNLP 2016 PBMT : Pourdamaghani and Knight, INLG 2016

  31. First Attempt TreeToStr TSP PBMT NNLG 29 26.9 23.2 23 22.4 17.4 BLEU 11.6 5.8 0 TreeToStr : Flanigan et al, NAACL 2016 TSP : Song et al, EMNLP 2016 PBMT : Pourdamaghani and Knight, INLG 2016

  32. First Attempt TreeToStr TSP PBMT NNLG 29 26.9 23.2 23 22.4 22 17.4 BLEU 11.6 5.8 0 TreeToStr : Flanigan et al, NAACL 2016 TSP : Song et al, EMNLP 2016 PBMT : Pourdamaghani and Knight, INLG 2016

  33. First Attempt TreeToStr TSP PBMT NNLG 29 All systems use a 26.9 23.2 Language Model 23 22.4 22 trained on a very 17.4 large corpus. BLEU 11.6 We will emulate via data augmentation . 5.8 0 TreeToStr : Flanigan et al, NAACL 2016 TSP : Song et al, EMNLP 2016 (Sennrich et al., ACL 2016) PBMT : Pourdamaghani and Knight, INLG 2016

  34. What went wrong? hold Reference :ARG0 (person :ARG0-of (have-role US officials held an expert group meeting in January :ARG1 loc_0 2002 in New York . :ARG2 official) ) :ARG1 (meet Prediction :ARG0 (person :ARG1-of expert :ARG2-of group) United States officials held held a meeting in ) January 2002 . :time (date-entity year_0 month_0) :location loc_1 74.85% 44.26%

  35. What went wrong? hold Reference :ARG0 (person :ARG0-of (have-role US officials held an expert group meeting in January :ARG1 loc_0 2002 in New York . :ARG2 official) ) :ARG1 (meet Prediction :ARG0 (person :ARG1-of expert :ARG2-of group) United States officials held held a meeting in ) January 2002 . :time (date-entity year_0 month_0) :location loc_1 ‣ Repetition 74.85% 44.26%

  36. What went wrong? hold Reference :ARG0 (person :ARG0-of (have-role US officials held an expert group meeting in January :ARG1 loc_0 2002 in New York . :ARG2 official) ) :ARG1 (meet Prediction :ARG0 (person :ARG1-of expert :ARG2-of group) United States officials held held a meeting in ) January 2002 . :time (date-entity year_0 month_0) :location loc_1 ‣ Repetition ‣ Coverage 74.85% 44.26%

  37. What went wrong? hold Reference :ARG0 (person :ARG0-of (have-role US officials held an expert group meeting in January :ARG1 loc_0 2002 in New York . :ARG2 official) ) :ARG1 (meet Prediction :ARG0 (person :ARG1-of expert :ARG2-of group) United States officials held held a meeting in ) January 2002 . :time (date-entity year_0 month_0) :location loc_1 Total OOV@1 OOV@5 ‣ Repetition ‣ Coverage 18000 a) Sparsity 13500 Tokens 74.85% 9000 44.26% 4500 0

  38. What went wrong? hold Reference :ARG0 (person :ARG0-of (have-role US officials held an expert group meeting in January :ARG1 loc_0 2002 in New York . :ARG2 official) ) :ARG1 (meet Prediction :ARG0 (person :ARG1-of expert :ARG2-of group) United States officials held held a meeting in ) January 2002 . :time (date-entity year_0 month_0) :location loc_1 Total OOV@1 OOV@5 ‣ Repetition ‣ Coverage 18000 a) Sparsity 13500 Tokens b) Avg sent length: 20 words 74.85% 9000 c) Limited Language 44.26% Modeling capacity 4500 0

  39. Data Augmentation Original Dataset: ~16k graph-sentence pairs

  40. Data Augmentation Original Dataset: ~16k graph-sentence pairs Gigaword: ~183M sentences *only*

  41. Data Augmentation Original Dataset: ~16k graph-sentence pairs Gigaword: ~183M sentences *only* Sample sentences with vocabulary overlap Original Giga-200k Giga-2M 80 60 % 40 20 0 OOV@1 OOV@5

  42. Data Augmentation graph Generate from MR graph text Encoder Decoder text Attention

  43. Data Augmentation graph Parse to MR Generate from MR text graph graph text Encoder Decoder Encoder Decoder text Attention Attention

  44. Data Augmentation graph Parse to MR Generate from MR text graph graph text Encoder Decoder Encoder Decoder text Attention Attention

  45. Data Augmentation graph Re-train Parse to MR Generate from MR text graph graph text Encoder Decoder Encoder Decoder text Attention Attention

  46. Data Augmentation input Generate Parse to Input from Input text

  47. Paired Training

  48. Paired Training ( , ) Train MR Parser P on Original Dataset

  49. Paired Training ( , ) Train MR Parser P on Original Dataset for i = 0 … N

  50. Paired Training ( , ) Train MR Parser P on Original Dataset for i = 0 … N S i =Sample k 10 i sentences from Gigaword

  51. Paired Training ( , ) Train MR Parser P on Original Dataset for i = 0 … N S i =Sample k 10 i sentences from Gigaword Parse S i sentences with P

  52. Paired Training ( , ) Train MR Parser P on Original Dataset for i = 0 … N Self-train Parser S i =Sample k 10 i sentences from Gigaword Parse S i sentences with P Re-train MR Parser P on S i

  53. Paired Training ( , ) Train MR Parser P on Original Dataset for i = 0 … N Self-train Parser S i =Sample k 10 i sentences from Gigaword Parse S i sentences with P Re-train MR Parser P on S i

  54. Paired Training ( , ) Train MR Parser P on Original Dataset for i = 0 … N Self-train Parser S i =Sample k 10 i sentences from Gigaword Parse S i sentences with P Re-train MR Parser P on S i ( , ) Train Generator G on S N

  55. Training MR Parser Train P on Original Dataset

  56. Training MR Parser Train P on Original Dataset

  57. Training MR Parser Sample S 1 = 200k Train P on sentences Original Dataset from Gigaword 200k

  58. Training MR Parser Sample S 1 = 200k Train P on sentences Original Dataset from Gigaword ( , ) Parse S 1 with P 200k 200k

  59. Training MR Parser Sample S 1 = 200k Train P on sentences Original Dataset from Gigaword ( , ) Parse S 1 with P Train P on S 1 = 200k 200k 200k 200k

  60. Fine-tune : init parameters from previous step and train on Original Dataset Training MR Parser Sample S 1 = 200k Train P on sentences Original Dataset from Gigaword Fine-tune P on ( , ) Parse S 1 with P Original Dataset Train P on S 1 = 200k 200k 200k 200k

  61. Fine-tune : init parameters from previous step and train on Original Dataset Training MR Parser Sample S 2 = 2M sentences from Gigaword Fine-tune P on ( , ) Parse S 2 with P Original Dataset Train P on S 2 = 2M 200k 200k 200k

  62. Fine-tune : init parameters from previous step and train on Original Dataset Training MR Parser Sample S 2 = 2M sentences from Gigaword Fine-tune P on ( , ) Parse S 2 with P Original Dataset Train P on S 2 = 2M 2M 2M 200k 2M

  63. Fine-tune : init parameters from previous step and train on Original Dataset Training MR Generator Sample S 3 = 2M sentences from Gigaword Fine-tune G on ( , ) Parse S 3 with P Original Dataset Train G on S 3 = 2M 2M 2M 200k 2M

  64. Fine-tune : init parameters from previous step and train on Original Dataset Training MR Generator Sample S 3 = 2M sentences from Gigaword Fine-tune G on ( , ) Parse S 3 with P Original Dataset Train G on S 3 = 2M 2M 2M 2M G G 2M

  65. Fine-tune : init parameters from previous step and train on Original Dataset Training MR Generator Sample S 3 = 2M sentences from Gigaword Fine-tune G on ( , ) Parse S 3 with P Original Dataset Train G on S 3 = 2M 2M 2M 2M G G 2M

  66. Final Results TreeToStr : Flanigan et al, NAACL 2016 TSP : Song et al, EMNLP 2016 PBMT : Pourdamaghani and Knight, INLG 2016

  67. Final Results TreeToStr TSP PBMT NNLG NNLG-200k NNLG-2M NNLG-20M 35 28 26.9 23 21 22.4 22 BLEU 14 7 0 TreeToStr : Flanigan et al, NAACL 2016 TSP : Song et al, EMNLP 2016 PBMT : Pourdamaghani and Knight, INLG 2016

Recommend

Building Ubiquitous and Robust Speech and Natural Language I nterfaces I Gary Geunbae Lee,

Building Ubiquitous and Robust Speech and Natural Language I nterfaces I Gary Geunbae Lee,

Building Ubiquitous and Robust Speech and Natural Language I nterfaces I Gary Geunbae Lee, Ph.D., Professor Dept. CSE, POSTECH Contents PART-I: Statistical Speech/Language Processing (60min) Natural Language Processing short intro

1.33k views • 129 slides
Natural Language Understanding We want to communicate with computers using natural language

Natural Language Understanding We want to communicate with computers using natural language

Natural Language Understanding We want to communicate with computers using natural language (spoken and written). unstructured natural language allow any statements, but make mistakes or failure. controlled natural language only

698 views • 20 slides
Advanced Natural Language Processing: What is Natural Language Processing (NLP)? Background

Advanced Natural Language Processing: What is Natural Language Processing (NLP)? Background

Course Logistics What is Natural Language Processing? computers using natural language as input and/or Instructor Regina Barzilay, Michael Collins output Email regina@csail.mit.edu, mcollins@csail.mit.edu computer language language

552 views • 13 slides
STS for NLG Christian Chiarcos chiarcos@uni-potsdam.de Natural Language Generation Natural

STS for NLG Christian Chiarcos chiarcos@uni-potsdam.de Natural Language Generation Natural

STS for NLG Christian Chiarcos chiarcos@uni-potsdam.de Natural Language Generation Natural Language Generation (NLG) (...) is a subfield of artificial intelligence and computational linguistics that is concerned with building computer

557 views • 10 slides
Natural language is a programming language: Applying natural language processing to software

Natural language is a programming language: Applying natural language processing to software

Natural language is a programming language: Applying natural language processing to software development Michael D. Ernst Presented by: Tomas Geffner, Subendhu Rongali &amp; Natcha Simsiri Before we start, what is software? Not just code/AST

437 views • 23 slides
Building Natural Language System based on Theoretical Linguistics

Building Natural Language System based on Theoretical Linguistics

Building Natural Language System based on Theoretical Linguistics @MiCS 2019/10/23 Masashi Yoshikawa (NAIST D3) Univ.), and now back in Nara Self Introduction @Kuwait 2012 NAIST

1.1k views • 67 slides
Natural Language Processing George Konidaris gdk@cs.brown.edu Fall 2019 Natural Language

Natural Language Processing George Konidaris gdk@cs.brown.edu Fall 2019 Natural Language

Natural Language Processing George Konidaris gdk@cs.brown.edu Fall 2019 Natural Language Processing Understanding spoken/written sentences in a natural language. Major area of research in AI. Why? Humans use language to communicate.

1.17k views • 33 slides
Scalable Certification for Scalable Certification for Typed Assembly Language Typed Assembly

Scalable Certification for Scalable Certification for Typed Assembly Language Typed Assembly

Scalable Certification for Scalable Certification for Typed Assembly Language Typed Assembly Language Dan Grossman (with Greg Morrisett) Cornell University 2000 ACM SIGPLAN Workshop on Types in Compilation AFTER AF Types After After

253 views • 22 slides
Natural Language Processing Stages in understanding natural language Why its hard

Natural Language Processing Stages in understanding natural language Why its hard

Natural Language Processing Stages in understanding natural language Why its hard 14-11-2011 (Some slides adapted from presentations by Dan Jurafsky and Bonnie Dorr.) Computer Speech and Language Processing What is it? Getting

604 views • 11 slides
Neural Language Models The New Frontier of Natural Language Understanding Gabriele Sarti

Neural Language Models The New Frontier of Natural Language Understanding Gabriele Sarti

Neural Language Models The New Frontier of Natural Language Understanding Gabriele Sarti University of Trieste, SISSA &amp; ItaliaNLP Lab StaTalk 2019 Table of Contents Natural Language Processing: On a Quest for Meaning Modeling Natural

773 views • 22 slides
NLTK: The Natural Language Toolkit Edward Loper Natural Language Processing Use

NLTK: The Natural Language Toolkit Edward Loper Natural Language Processing Use

NLTK: The Natural Language Toolkit Edward Loper Natural Language Processing Use computational methods to process human language. Examples: Machine translation Text classification Text summarization Question answering

502 views • 29 slides
Pragmatic aspects of natural language Vojtch Kov Natural Language Processing Centre

Pragmatic aspects of natural language Vojtch Kov Natural Language Processing Centre

Outline Pragmatics Grices cooperative principle Speech acts Pragmatic aspects of natural language Vojtch Kov Natural Language Processing Centre Faculty of Informatics, Masaryk University Botanick 68a, 602 00 Brno

260 views • 7 slides
Outline of todays lecture Overview of Natural Language Generation Components of Natural

Outline of todays lecture Overview of Natural Language Generation Components of Natural

Natural Language Processing Outline of todays lecture Overview of Natural Language Generation Components of Natural Language Generation systems Data for NNs via classical realization Referring expressions Natural Language Processing

586 views • 37 slides
LANGUAGE MODELS 24.05.19 Statistical Natural Language Processing 1 Statistical natural

LANGUAGE MODELS 24.05.19 Statistical Natural Language Processing 1 Statistical natural

Jurafsky, D. and Martin, J. H. (2009): Speech and Language Processing. An Introduction to Natural Language Processing , Computational Linguistics and Speech Recognition. Second Edition. Pearson: New Jersey. Chapter 4 Manning, C. D. and

898 views • 60 slides
Natural Language Interaction Gurpreet Singh Papers Learning to Parse Natural Language

Natural Language Interaction Gurpreet Singh Papers Learning to Parse Natural Language

Natural Language Interaction Gurpreet Singh Papers Learning to Parse Natural Language Commands to a Robot Control System (Cynthia Matuszek, Evan Herbst, Luke Zettlemoyer, Dieter Fox) Interpreting Instruction Sequences in Spatial

445 views • 29 slides
Opening the pod bay doors building intelligent agents that can interpret, generate and learn

Opening the pod bay doors building intelligent agents that can interpret, generate and learn

Opening the pod bay doors building intelligent agents that can interpret, generate and learn from natural language Jacob Andreas, MIT / Microsoft web.mit.edu/jda/www / @jacobandreas Following natural language instructions

1.61k views • 122 slides
1 NATURAL LOGIC IN NATURAL LANGUAGE Johan van Benthem,

1 NATURAL LOGIC IN NATURAL LANGUAGE Johan van Benthem,

1 NATURAL LOGIC IN NATURAL LANGUAGE Johan van Benthem, http://staff.fnwi.uva.nl/j.vanbenthem ILLC, Logic, Language and Computation, Monday 8 Sept 2014 Abstract

305 views • 4 slides
Natural Language Processing with Deep Learning CS224N/Ling284 Lecture 15: Natural Language

Natural Language Processing with Deep Learning CS224N/Ling284 Lecture 15: Natural Language

Natural Language Processing with Deep Learning CS224N/Ling284 Lecture 15: Natural Language Generation Christopher Manning Announcements Thank you for all your hard work! We know Assignment 5 was tough and a real challenge to do

1.24k views • 87 slides
Advanced Natural Language Processing: What is Natural Language Processing (NLP)? Background

Advanced Natural Language Processing: What is Natural Language Processing (NLP)? Background

Overview Advanced Natural Language Processing: What is Natural Language Processing (NLP)? Background and Overview Why is NLP hard? What will this course be about? Michael Collins EECS/CSAIL September 6, 2007 Advanced Natural

296 views • 7 slides
Let the AI do the Talk Adventures with Natural Language Generation @MarcoBonzanini PyParis 2018

Let the AI do the Talk Adventures with Natural Language Generation @MarcoBonzanini PyParis 2018

Let the AI do the Talk Adventures with Natural Language Generation @MarcoBonzanini PyParis 2018 PyData London Conference 12-14 July 2019 @PyDataLondon NATURAL LANGUAGE GENERATION Natural Language Processing 5 Natural Language Processing

1.43k views • 84 slides
Information Extraction Industrial Natural Language Processing Industrial Natural Language

Information Extraction Industrial Natural Language Processing Industrial Natural Language

Industrial Natural Language Processing &amp; Information Extraction Industrial Natural Language Processing Industrial Natural Language Processing Overview Natural Language Processing Developing and applying techniques NLP and methods for

479 views • 20 slides
Introduction to Natural Language Processing CMSC 470 Marine Carpuat Natural Language Processing

Introduction to Natural Language Processing CMSC 470 Marine Carpuat Natural Language Processing

Introduction to Natural Language Processing CMSC 470 Marine Carpuat Natural Language Processing (NLP) The engineering discipline of doing what people do with language, but using computers Computational Linguistics (CL) The science of

934 views • 28 slides
Statistical Natural Language Processing Prasad Tadepalli CS430 lecture Natural Language

Statistical Natural Language Processing Prasad Tadepalli CS430 lecture Natural Language

Statistical Natural Language Processing Prasad Tadepalli CS430 lecture Natural Language Processing Some subproblems are partially solved Spelling correction, grammar checking Information retrieval with keywords Semi-automatic

483 views • 9 slides
Building Fair and Robust Representations for Vision and Language Vicente Ordez-Romn

Building Fair and Robust Representations for Vision and Language Vicente Ordez-Romn

EMNLP 2019 Tutorial on Bias and Fairness in Natural Language Processing, Hong Kong Building Fair and Robust Representations for Vision and Language Vicente Ordez-Romn Assistant Professor Department of Computer Science Outline Issues

739 views • 60 slides

More recommend