Deep learning trends
Download as PPTX, PDF0 likes89 views
This document contains summaries of several topics: 1) It discusses deep learning and how stacking many simple functions can lead to complex models for tasks like image classification. 2) It notes some growing trends in deep learning research like an increasing number of papers being submitted to top conferences. 3) It lists some major companies investing heavily in deep learning research like Google, Facebook, Microsoft and Amazon. 4) It provides an overview of the author's research which aims to synthesize optical flow from pre-trained disparity estimation networks since optical flow data can be difficult to obtain.
Deep learning trends
- 1. Disp2Flow: Optical flow synthesis using pre-trained disparity estimation networks 전기정보공학부 김준호
- 2. Disp2Flow: Optical flow synthesis using pre-trained disparity estimation networks 전기정보공학부 김준호
- 3. Deep learning trends + My Research 전기정보공학부 김준호
- 4. Contents • What is deep learning? • Growing trends in deep learning • Big players in deep learning • Deep learning @Korea • My research
- 5. What is deep learning?
- 6. What is deep learning?
- 7. What is machine learning? • Data -> Learn -> Knowledge (Make predictions on unseen data) • Examples Support Vector Machines Gaussian Process Regression
- 8. What is machine learning?(cont.)
- 9. FAQ: How does unsupervised learning work?
- 10. Unsupervised learning is cool but… • In many cases some form of supervision is necessary • “Fully unsupervised” learning is impossible -> Some assumptions about input data should be given! • No free lunch theorem • Let’s only focus on supervised learning for now!
- 11. Supervised learning • Goal: Find a function “f” that maps input data to output domain • Input & output domain could be anything • Examples • Input: Images -> Output: Label • Input: Korean -> Output: English
- 12. What is deep learning? • Consider a supervised learning problem where we want to predict whether an image is a cat or not • Again, we want to find “f” • The relationship between an image and its label will be very complex • Probably not “y = ax + b” • We can build “complex” functions using composition! • Stack “a lot” of functions -> deep learning
- 13. What is deep learning? Y = f(a*f(c*f(e*f….(x)+g)+d) + b)
- 14. Growing(Bad) trends in deep learning
- 15. Hmmm…
- 16. Major AI research areas
- 17. Major AI research areas
- 18. Top-tier machine learning conferences • Computer vision • CVPR • ICCV • ECCV • Natural language processing • ACL • EMNLP • NAACL-HLT • Machine learning • NIPS • ICML • ICLR
- 19. Number of submissions • CVPR – top tier conference in computer vision • Acceptance rate: approx. 23%
- 20. Number of submissions • NIPS – top tier machine learning conference • Acceptance rate: approx. 20%
- 21. A typical review process • Most machine learning conferences use a peer-review system • Papers are submitted to different “areas” • Ex) If one submits a paper on classifying cats, it will go under the category of “image classification” • Each “area” has an “area chair” – a person who is in charge of all submission to that “area” • “Area chairs” distribute the submitted papers to reviewers, who will eventually decide accept/reject
- 22. More papers are good but… • It means we need more reviewers • However, the total number of reviewers is limited • A lot of bad quality reviewers are present
- 24. Big players in deep learning
- 25. = Diamond sponsors of top-tier conferences
- 26. CVPR
- 27. ACL
- 28. Ilya Sutskever • Research scientist @OpenAI • Salary - $1.9 million
- 29. Ian Goodfellow • Research scientist @ Google brain • Salary - $800,000
- 31. Number of top-tier conference papers are increasing! • CVPR 2017, 2018 -> More than 20 papers accepted • NIPS 2018 -> 20 papers accepted (~5 in NIPS 2017) • SNU ECE: 3 • SNU CS: 1 • SNU 융기원: 2 • KAIST EE: 5 • KAIST CS: 3 • Yonsei EE: 1 • Yonsei CS: 1 • Kakao brain: 1 • NAVER: 1 • SK T-brain: 1 • Lunit: 1
- 32. “Four” companies
- 33. My research
- 34. Disparity • Finding disparity leads to depth estimation in stereo vision • In order to find disparity, we need rectified stereo pairs • Matching points should be on the ‘same’ horizontal line
- 35. Rectified images
- 36. Optical flow • Optical flow is a more general problem, where we want to find individual movements of pixels
- 37. Optical flow -> Disparity • A 90s paper
- 38. Disparity -> Optical flow • Optical flow training data is hard to find • In many cases synthetic data is used • Could we use pre-trained models for disparity estimation to find optical flow?
- 39. Disparity -> Optical flow • Input: Image pair • Output: Optical flow • Novelty: Obtain optical flow estimation using networks trained on stereo disparity estimation • Method overview • Pre-train a neural network(DispNet-Horizontal) on stereo disparity data • Using basic rotation operations, obtain another neural network(DispNet- Vertical) on rotated stereo disparity data • Train a domain transfer network to map optical flow target images to stereo images(automatic rectification) • Fine-tune domain transfer network using optical flow ground truth
- 40. Disparity -> Optical flow (initial version)
- 41. Disparity -> Optical flow (new version)
- 42. Disparity -> Optical flow(TODO) • Pre-train dispnet • Download optical flow dataset • Test adequate volume for discriminator • Implement filter rotation
- 43. Questions?