View Project

Title:

Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders

Abstract:

While recent neural encoder-decoder models have shown great promise in modeling open-domain conversations, they often generate dull and generic responses. Unlike past work that has focused on diversifying the output of the decoder at word-level to alleviate this problem, we present a novel framework based on conditional variational autoencoders that captures the discourse-level diversity in the encoder. Our model uses latent variables to learn a distribution over potential conversational intents and generates diverse responses using only greedy decoders. We have further developed a novel variant that is integrated with linguistic prior knowledge for better performance. Finally, the training procedure is improved by introducing a bag-of-word loss. Our proposed models have been validated to generate significantly more diverse responses than baseline approaches and exhibit competence in discourse-level decision-making.

Comments:

Actions

Suggested Topics

Full Matches (full topic name in abstract)

Partial Matches (at least half of words topic name appear in abstract)

Suggested Resources

Uses abstract to search the content of resources available in Topics. Sorted by relevance.

# Title Author Topic Medium Score
1 Recent Advances in Document Summarization Jin-ge Yao, Xiaojun Wan, Jianguo Xiao 421 survey 153.93
2 The 7 NLP Techniques That Will Change How You Communicate in the Future (Part I) James Le 112 resource 144.74
3 Deep Learning for NLP: An Overview of Recent Trends Elvis 711 resource 141.23
4 Neural information retrieval: at the end of the early years Kezban Dilek Onal, Ye Zhang, Ismail Sengor Altingovde, Md Mustafizur Rahman, Pinar Karagoz, Alex Braylan, Brandon Dang, Heng-Lu Chang, Henna Kim, Quinten McNamara, Aaron Angert, Edward Banner, Vivek K 232 survey 140.91
5 DEEP LEARNING FOR CHATBOTS, PART 2 - IMPLEMENTING A RETRIEVAL-BASED MODEL IN TENSORFLOW Denny Britz 445 tutorial 138.41
6 ICML+ACL’18: Structure Back in Play, Translation Wants More Context Andre Martins 956 resource 137.40
7 Codra: A Novel Discriminative Framework for Rhetorical Analysis Shafiq Joty, Giuseppe Carenini, Raymond T. Ng 999 paper 132.09
8 DEEP LEARNING FOR CHATBOTS, PART 1 - INTRODUCTION Denny Britz 445 tutorial 129.34
9 The 7 NLP Techniques That Will Change How You Communicate in the Future (Part II) James Le 133 resource 125.22
10 Summaries and notes on Deep Learning research papers Denny Britz 713 resource 122.52
11 Chatsbots with Machine Learning: Building Neural Conversational Agents Dmitry Persiyanov 999 resource 122.35
12 The Current Best of Universal Word Embeddings and Sentence Embeddings Thomas Wolf 721 resource 122.20
13 NLP’s generalization problem, and how researchers are tackling it Ana Marasovic 711 resource 122.18
14 A Comparative Analysis of ChatBots APIs Author Unknown 921 resource 119.41
15 Using Artificial Intelligence to Augment Human Intelligence Shan Carter, Michael Nielsen 811 resource 118.37
16 A survey of transfer learning Karl Weiss, Taghi M. Khoshgoftaar and DingDing Wang 978 resource 116.06
17 Deep Learning Summer School Invited Speakers with slides CIFAR 713 resource 114.75
18 Diversity and network coherence as indicators of interdisciplinarity: case studies in bionanoscience Ismael Rafols, Martin Meyer 999 paper 113.61
19 Highlights of EMNLP 2017: Exciting Datasets, Return of the Clusters, and More! Sebastian Ruder 641 resource 112.85
20 End-to-end speech Anthony Ndirango 863 resource 112.44
21 Building Autoencoders in Keras Francois Chollet 731 tutorial 112.05
22 Similarity-driven Semantic Role Induction via Graph Partitioning Joel Lang, Mirella Lapata 999 paper 111.51
23 Transfer Learning - Machine Learnings Next Frontier Sebastian Ruder 978 tutorial 111.13
24 Deep Learning for NLP Best Practices Sebastian Ruder 713 tutorial 110.94
25 Discriminative Syntax-based Word Ordering for Text Generation Yue Zhang, Stephen Clark 999 paper 110.86
26 ACL 2017 Report Yuta Kikuchi, Sosuke Kobayashi 711 resource 110.29
27 A Gentle Introduction to Neural Networks for Machine Learning James Le 711 resource 110.21
28 Neural Text Embeddings for IR Bhaskar Mitra, Nick Craswell 721 tutorial 110.18
29 The 10 Neural Network Architectures Machine Learning Researchers Need To Learn James Le 641 resource 110.04
30 What I learned from Deep Learning Summer School 2016 Hamid Palangi 107 tutorial 109.41
31 Neural text generation: How to generate text using conditional language models Neil Yager 43 resource 109.18
32 Deep Learning for NLP, advancements and trends in 2017 Javier 711 resource 108.52
33 The Unreasonable Ineffectiveness of Deep Learning in NLU Suman Deb Roy 713 tutorial 108.32
34 Your tl;dr by an ai: a deep reinforced model for abstractive summarization Romain Paulus 754 tutorial 108.01
35 Neural Machine Translation (seq2seq) Tutorial Thang Luong, Eugene Brevdo, Rui Zhao 753 tutorial 107.86
36 Machine Translation Without the Data Harshvardhan Gupta 451 resource 107.52
37 NLP's ImageNet moment has arrived Sebastian Ruder 862 resource 107.38
38 Rohan & Lenny #3: Recurrent Neural Networks & LSTMs Rohan Kapur 741 tutorial 107.29
39 Your tl;dr by an Ai: a Deep Reinforced Model for Abstractive Summarization Romain Paulus 811 resource 106.83
40 Adversarial Autoencoders (with Pytorch) Felipe N. Ducau 713 tutorial 106.72
41 Deep Learning chatbots analysis - whats the actual tech behind them? Przemyslaw 756 tutorial 106.55
42 Tombones Computer Vision Blog Tomasz Malisiewicz 958 resource 105.90
43 Generating Large Images from Latent Vectors Author Unknown 631 tutorial 105.38
44 A survey of cross-lingual embedding models Sebastian Ruder 721 tutorial 104.76
45 Towards data set augmentation with GANs Pedro Ferreira 713 resource 104.40
46 What is the difference between convolutional neural networks, restricted Boltzmann machines, and auto-encoders? ffriend 744 tutorial 104.11
47 Taming Recurrent Neural Networks for Better Summarization Abigail See 754 tutorial 103.77
48 Generative Model Chatbots Kumar Shridhar 756 tutorial 103.57
49 Deconstruction with Discrete Embeddings R2RT 711 resource 103.30
50 Building a FAQ Chatbot in Python – The Future of Information Searching Yogesh Kulkarni 232 resource 103.10
51 Negated bio-events: analysis and identification Raheel Nawaz, Paul Thompson, Sophia Ananiadou 999 paper 102.71
52 Multi-Task Learning Objectives for Natural Language Processing Author Unknown 133 resource 102.47
53 Improving Language Understanding with Unsupervised Learning Alec Radford 581 resource 101.80
54 Deep Generative Models Prakash Pandey 711 resource 101.15
55 Generative Adversarial Networks (GANs): Engine and Applications Anton Karazeev 713 resource 101.03
56 Learned in translation: contextualized word vectors Bryan McCann 452 resource 100.75
57 Word embeddings in 2017: Trends and future directions Sebastian Ruder 721 resource 100.61
58 A miscellany of fun deep learning papers Adrian Colyer 711 resource 100.47
59 Deep Learning Achievements Over the Past Year Eduard Tyantov 711 resource 100.46
60 Introduction to Neural Machine Translation with GPUs (part 3) Kyunghyun Cho 753 tutorial 100.18
61 Minibatch Metropolis-Hastings Daniel Seita 107 tutorial 99.94
62 State-of-the-art neural coreference resolution for chatbots Thomas Wolf 756 tutorial 99.73
63 Advances in Semantic Textual Similarity Yinfei Yang, Chris Tar 311 resource 99.72
64 Ideas on interpreting machine learning Patrick Hall, Wen Phan, SriSatish Ambati 134 tutorial 99.62
65 Tutorial - What is a variational autoencoder? Jaan Altosaar 711 tutorial 99.38
66 Deep Learning in NLP Vered Shwartz 711 resource 98.55
67 Deep Learning Transcends the Bag of Words Zachary Lipton 741 tutorial 98.53
68 Four deep learning trends from ACL 2017: Part 1 Abigail See 713 resource 97.93
69 Recurrent Neural Networks Stephen Grossberg 741 paper 97.89
70 The Illustrated Transformer Jay Alammar 745 resource 97.80
71 An Overview of Multi-Task Learning in Deep Neural Networks Sebastian Ruder 829 tutorial 97.76
72 The 8 Neural Network Architectures Machine Learning Researchers Need to Learn James Le' 731 resource 97.71
73 Fully-Parallel Text Generation for Neural Machine Translation Jiatao Gu, James Bradbury 711 resource 97.45
74 Lenny #2: Autoencoders and Word Embeddings Lenny Khazan 721 tutorial 97.29
75 A social network's changing statistical properties and the quality of human innovation Brian Uzzi 999 paper 97.12