1 / 11

-Jason Weston, Frederic Ratle , Hossein Mobahi and Ronan Collobert Google , New York, USA.

-Jason Weston, Frederic Ratle , Hossein Mobahi and Ronan Collobert Google , New York, USA. Review by A.U.S.S Pradeep. What is Deep Learning?. “Deep learning” is the new big trend in Machine Learning. It promises general, powerful, and fast machine learning, moving us one step closer to AI.

creola
Download Presentation

-Jason Weston, Frederic Ratle , Hossein Mobahi and Ronan Collobert Google , New York, USA.

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. -Jason Weston, Frederic Ratle, HosseinMobahiand Ronan CollobertGoogle, New York, USA. Review by A.U.S.S Pradeep

  2. What is Deep Learning? • “Deep learning” is the new big trend in Machine Learning. It promises general, powerful, and fast machine learning, moving us one step closer to AI. • Deep learning algorithms attempt to learn multiple levels of representation of increasing complexity/abstraction. The general architecture of Deep learning algorithm involves- • Extract low level features. • Extract more complicated features as we progress called pre-training. • Perform supervise task at the end, and fine tune the weights by back propagation.

  3. Shallow Vs Deep methods Deep Researchers (DRs) believe: • Learn sub-tasks in layers. Essential for hard tasks. • Natural for multi-task learning. • Non-linearity is efficient compared to shallow methods. Shallow Researchers believe: • NNs were already complicated and messy. • New deep methods are even more complicated and messy. • Shallow methods: clean and give valuable insights into what works. So why not try to borrow the nice elements of shallow methods and put it in deep learning framework ?

  4. Existing Embedding Algorithms Many existing (“shallow”) embedding algorithms optimize:min MDS: minimize ISOMAP: same equation, but W defined by shortest path on neighborhood graph. LapSVM [Belkin et al.]: SVM+ = 1 if two points are neighbours , 0 otherwise.

  5. Semi-supervised learning in Deep arhitecture • Define a neural network, • Supervised Training : to minimize • Can add unsupervised training to any of the layers, • Output: • Intermediate: • Auxiliary:

  6. Deep semi-supervised learning

  7. Results

  8. References: • Deep Learning forNLP (without Magic) , Socher. • weston-ratle-collobert-12_deep-learning-via-semi-supervised-embedding. • Icml09 Weston dlss 01 • Icml08 ratledls 01

  9. Thank you !! Questions ?

  10. Part-Of-Speech Tagging (POS): syntactic roles (noun, adverb...) • Chunking: syntactic constituents (noun phrase, verb phrase...) • Name Entity Recognition (NER): person/company/location... • Semantic Role Labeling (SRL): semantic role

More Related