Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Sebastian Ruder

Research Scientist, AYLIEN
PhD Candidate, Insight Centre
@seb_ruder |01.03.17 | LinkedIn Tech Talk
Transf...
Agenda
1. What is Transfer Learning?
2. Why Transfer Learning now?
3. Transfer Learning in practice
4. Transfer Learning f...
What is Transfer Learning?
@seb_ruder |01.03.17 | LinkedIn Tech Talk
Model A Model B
Task / domain A
Task / domain B
Tradi...
What is Transfer Learning?
@seb_ruder |
Knowledge
Model
Source task /
domain Target task /
domain
Transfer learning
Storin...
@seb_ruder |
“Transfer learning
will be the next
driver of ML
success.”
Andrew Ng,
NIPS 2016 keynote
@seb_ruder |
Why Transfer Learning now?
@seb_ruder |
Supervised learning
Transfer learning
Unsupervised learning
Reinforcement learning...
Why Transfer Learning now?
@seb_ruder |
1. Learn very accurate input-output mapping
2. Maturity of ML models
- Computer vi...
Transfer Learning in practice
@seb_ruder |
• Train new model on features
of large model trained on
ImageNet3
• Train model...
Transfer Learning in practice
@seb_ruder |
• Progressive Neural

Networks7 have
access to weights
from trained models
• Pa...
Transfer Learning for NLP
@seb_ruder |
• Task and domainT D
DS 6= DT TS 6= TT
A (slightly) more technical definition
• Doma...
Transfer Learning for NLP
@seb_ruder |
Transfer scenarios
1. : Different topics, text types, etc.

2. : Different language...
Transfer Learning for NLP
@seb_ruder |
Current status
• Not as straightforward as in CV
- No universal deep features
• How...
Our research
@seb_ruder |
Research focus
Finding better ways to transfer knowledge to new
domains, tasks, and languages th...
Our research
@seb_ruder |
Training and test distributions are different.
Different text types. Different accents/ages.
Dif...
Our research
@seb_ruder |
Transfer learning challenges in real-world applications
1. Domains are not well-defined, but fuzz...
Our research
@seb_ruder |
• Idea: Use distillation + insights from semi-supervised
learning to transfer knowledge from a s...
Our research
@seb_ruder |
• Idea: Take into account diversity of training data to
select subsets (c) rather than an entire...
Our research
@seb_ruder |
Opportunities and future directions
• Learn from past adaptation scenarios and
generalise across...
References
@seb_ruder |
Image credit
• Google Research blog post11
• Mikolov, T., Joulin, A., & Baroni, M. (2015). A Roadm...
@seb_ruder |
Thanks for your attention!
Questions?
01.03.17 | LinkedIn Tech Talk
Upcoming SlideShare
Loading in …5
×

Transfer Learning -- The Next Frontier for Machine Learning

2,170 views

Published on

Tech talk on transfer learning given at LinkedIn, Dublin on March 1, 2017.

Published in: Science
  • Be the first to comment

Transfer Learning -- The Next Frontier for Machine Learning

  1. 1. Sebastian Ruder
 Research Scientist, AYLIEN PhD Candidate, Insight Centre @seb_ruder |01.03.17 | LinkedIn Tech Talk Transfer Learning — The Next Frontier for ML
  2. 2. Agenda 1. What is Transfer Learning? 2. Why Transfer Learning now? 3. Transfer Learning in practice 4. Transfer Learning for NLP 5. Our research 6. Opportunities and directions @seb_ruder |01.03.17 | LinkedIn Tech Talk
  3. 3. What is Transfer Learning? @seb_ruder |01.03.17 | LinkedIn Tech Talk Model A Model B Task / domain A Task / domain B Traditional ML Training and evaluation on the same task or domain.
  4. 4. What is Transfer Learning? @seb_ruder | Knowledge Model Source task / domain Target task / domain Transfer learning Storing knowledge gained solving one problem and applying it to a different but related problem. Model 01.03.17 | LinkedIn Tech Talk
  5. 5. @seb_ruder |
  6. 6. “Transfer learning will be the next driver of ML success.” Andrew Ng, NIPS 2016 keynote @seb_ruder |
  7. 7. Why Transfer Learning now? @seb_ruder | Supervised learning Transfer learning Unsupervised learning Reinforcement learning 2016Time Commercial success Drivers of ML success in industry - Andrew Ng, NIPS 2016 keynote 01.03.17 | LinkedIn Tech Talk
  8. 8. Why Transfer Learning now? @seb_ruder | 1. Learn very accurate input-output mapping 2. Maturity of ML models - Computer vision (5% error on ImageNet) -Automatic speech recognition (3x faster than typing, 20% more accurate1) 3. Large-scale deployment & adoption of ML models -Google’s NMT System2 1 Ruan, S., Wobbrock, J. O., Liou, K., Ng, A., & Landay, J. (2016). Speech Is 3x Faster than Typing for English and Mandarin Text Entry on Mobile Devices. arXiv preprint arXiv:1608.07323. 2 Wu, Y., Schuster, M., Chen, Z., Le, Q. V, Norouzi, M., Macherey, W., … Dean, J. (2016). Google’s Neural Machine Translation System: Bridging the Gap between Human and Machine Translation. arXiv preprint arXiv:1609.08144. Huge reliance on labeled data 
 Novel tasks / domains without (labeled) data 01.03.17 | LinkedIn Tech Talk
  9. 9. Transfer Learning in practice @seb_ruder | • Train new model on features of large model trained on ImageNet3 • Train model to confuse source and target domains4 • Train model on domain- invariant representations5,6 3 Razavian, A. S., Azizpour, H., Sullivan, J., & Carlsson, S. (2014). CNN features off-the-shelf: An astounding baseline for recognition. IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, 512–519. 4 Ganin, Y., & Lempitsky, V. (2015). Unsupervised Domain Adaptation by Backpropagation. Proceedings of the 32nd International Conference on Machine Learning., 37. 5 Bousmalis, K., Trigeorgis, G., Silberman, N., Krishnan, D., & Erhan, D. (2016). Domain Separation Networks. NIPS 2016. 6 Sener, O., Song, H. O., Saxena, A., & Savarese, S. (2016). Learning Transferrable Representations for Unsupervised Domain Adaptation. NIPS 2016. Computer vision 01.03.17 | LinkedIn Tech Talk
  10. 10. Transfer Learning in practice @seb_ruder | • Progressive Neural
 Networks7 have access to weights from trained models • PathNet8 learns weight paths via a genetic algorithm 7 Rusu, A. A., Rabinowitz, N. C., Desjardins, G., Soyer, H., Kirkpatrick, J., Kavukcuoglu, K., … Deepmind, G. (2016). Progressive Neural Networks. arXiv preprint arXiv:1606.04671. 8 Fernando, C., Banarse, D., Blundell, C., Zwols, Y., Ha, D., Rusu, A. A., … Wierstra, D. (2017). PathNet: Evolution Channels Gradient Descent in Super Neural Networks. In arXiv preprint arXiv:1701.08734. Reinforcement learning 01.03.17 | LinkedIn Tech Talk
  11. 11. Transfer Learning for NLP @seb_ruder | • Task and domainT D DS 6= DT TS 6= TT A (slightly) more technical definition • Domain where - : feature space, e.g. BOW representations - : e.g. distribution over terms in documents D = {X, P(X)} X P(X) • Task where - : label space, e.g. true/false labels - : learned mapping from samples to labels T = {Y, P(Y |X)} Y P(Y |X) • Transfer learning:
 Learning when or 01.03.17 | LinkedIn Tech Talk
  12. 12. Transfer Learning for NLP @seb_ruder | Transfer scenarios 1. : Different topics, text types, etc.
 2. : Different languages.
 3. : Unbalanced classes.
 4. : Different tasks. P(XS) 6= P(XT ) XS 6= XT P(YS|XS) 6= P(YT |XT ) YS 6= YT 01.03.17 | LinkedIn Tech Talk
  13. 13. Transfer Learning for NLP @seb_ruder | Current status • Not as straightforward as in CV - No universal deep features • However: “Simple” transfer through word embeddings is pervasive • History of research for task-specific transfer, e.g. sentiment analysis, POS tagging leveraging NLP phenomena such as structured features, sentiment words, etc. • Few research on transfer between tasks • More recently: representation-based research 01.03.17 | LinkedIn Tech Talk
  14. 14. Our research @seb_ruder | Research focus Finding better ways to transfer knowledge to new domains, tasks, and languages that 1. perform well in large-scale settings and real- world applications; 2. are applicable to many tasks and models. Current focus: : Training and test distributions are different. P(XS) 6= P(XT ) 01.03.17 | LinkedIn Tech Talk
  15. 15. Our research @seb_ruder | Training and test distributions are different. Different text types. Different accents/ages. Different topics/categories. Performance drop or even collapse is inevitable. 01.03.17 | LinkedIn Tech Talk
  16. 16. Our research @seb_ruder | Transfer learning challenges in real-world applications 1. Domains are not well-defined, but fuzzy and conflate many factors.
 
 
 2. One-to-one adaptation is rare and many source domains are generally available. 3. Models need to be adapted frequently as conditions change, new data becomes available, etc. Language socialfactors genre topic 01.03.17 | LinkedIn Tech Talk
  17. 17. Our research @seb_ruder | • Idea: Use distillation + insights from semi-supervised learning to transfer knowledge from a single (a) and multiple teachers (b) to a student model9. (a) (b) 9 Ruder, S., Ghaffari, P., & Breslin, J. G. (2017). Knowledge Adaptation: Teaching to Adapt. In arXiv preprint arXiv:1702.02052. How to adapt from large source domains? 01.03.17 | LinkedIn Tech Talk
  18. 18. Our research @seb_ruder | • Idea: Take into account diversity of training data to select subsets (c) rather than an entire domain (a) or individual examples (b)10. 10 Ruder, S., Ghaffari, P., & Breslin, J. G. (2017). Data Selection Strategies for Multi-Domain Sentiment Analysis. In arXiv preprint arXiv:1702.02426. How to select data for adaptation? (a) (b) (c) 01.03.17 | LinkedIn Tech Talk
  19. 19. Our research @seb_ruder | Opportunities and future directions • Learn from past adaptation scenarios and generalise across domains and tasks. • Robust adaptation to non-English and low- resource languages. • Adaptation for novel tasks and more sophisticated models, e.g. QA and memory networks. • Transfer across tasks and leveraging knowledge from related tasks. 01.03.17 | LinkedIn Tech Talk
  20. 20. References @seb_ruder | Image credit • Google Research blog post11 • Mikolov, T., Joulin, A., & Baroni, M. (2015). A Roadmap towards Machine Intelligence. arXiv preprint arXiv:1511.08130. • Google Research blog post12 Our papers • Ruder, S., Ghaffari, P., & Breslin, J. G. (2017). Knowledge Adaptation: Teaching to Adapt. In arXiv preprint arXiv:1702.02052. • Ruder, S., Ghaffari, P., & Breslin, J. G. (2017). Data Selection Strategies for Multi-Domain Sentiment Analysis. In arXiv preprint arXiv: 1702.02426. 11 https://research.googleblog.com/2016/10/how-robots-can-acquire-new-skills-from.html 12 https://googleblog.blogspot.ie/2014/04/the-latest-chapter-for-self-driving-car.html 01.03.17 | LinkedIn Tech Talk
  21. 21. @seb_ruder | Thanks for your attention! Questions? 01.03.17 | LinkedIn Tech Talk

×