User contributions for Jgosmann

27 November 2015

• 11:5511:55, 27 November 2015 diff hist +2‎ m on the Number of Linear Regions of Deep Neural Networks ‎ Fix math typos
• 11:5511:55, 27 November 2015 diff hist +6‎ m on the Number of Linear Regions of Deep Neural Networks ‎ Fix math typos

25 November 2015

• 14:2014:20, 25 November 2015 diff hist +520‎ learning Phrase Representations ‎ Add section about representations.
• 14:1914:19, 25 November 2015 diff hist 0‎ N File:Fig5.png ‎No edit summary current
• 14:1814:18, 25 November 2015 diff hist 0‎ N File:Fig4.png ‎No edit summary current
• 12:5212:52, 25 November 2015 diff hist +29‎ f15Stat946PaperSignUp ‎ Set B paper numbers
• 12:5012:50, 25 November 2015 diff hist +1,018‎ extracting and Composing Robust Features with Denoising Autoencoders ‎ Add information theoretic perspective.

24 November 2015

• 12:5212:52, 24 November 2015 diff hist +1,106‎ scene Parsing with Multiscale Feature Learning, Purity Trees, and Optimal Covers Machines ‎ Add info on training

18 November 2015

• 16:0116:01, 18 November 2015 diff hist +7,778‎ N on the Number of Linear Regions of Deep Neural Networks ‎ Created page with "= Introduction = The paper basically seeks to answer the question why deep neural networks perform so much better than shallow neural networks. It is not obvious that deep neura..."
• 16:0116:01, 18 November 2015 diff hist 0‎ N File:Constr.png ‎No edit summary current
• 15:5815:58, 18 November 2015 diff hist 0‎ N File:Montifar2.png ‎No edit summary current
• 15:5515:55, 18 November 2015 diff hist 0‎ N File:Montifar3.png ‎No edit summary current
• 15:5315:53, 18 November 2015 diff hist 0‎ N File:Hyperplanes.png ‎No edit summary current
• 15:5215:52, 18 November 2015 diff hist 0‎ N File:Sets.png ‎No edit summary current
• 15:5115:51, 18 November 2015 diff hist 0‎ N File:Montifar1.png ‎No edit summary current

12 November 2015

• 17:0517:05, 12 November 2015 diff hist +1,236‎ from Machine Learning to Machine Reasoning ‎ Added how association/dissociation modules relate to discrete Lips equivalents. Mentions VSAs.
• 15:1915:19, 12 November 2015 diff hist −5‎ m question Answering with Subgraph Embeddings ‎ →‎Task Definition
• 15:1715:17, 12 November 2015 diff hist +4‎ m question Answering with Subgraph Embeddings ‎ →‎Task Definition
• 15:1615:16, 12 November 2015 diff hist +26‎ m question Answering with Subgraph Embeddings ‎ →‎Introduction
• 15:1515:15, 12 November 2015 diff hist +3‎ m question Answering with Subgraph Embeddings ‎No edit summary
• 14:5414:54, 12 November 2015 diff hist −19‎ f15Stat946PaperSignUp ‎No edit summary

10 November 2015

• 14:5214:52, 10 November 2015 diff hist +152‎ deep Sparse Rectifier Neural Networks ‎ Adding key result about pre-training.
• 14:4714:47, 10 November 2015 diff hist +235‎ deep Sparse Rectifier Neural Networks ‎ More benefits of sparse representation
• 14:4414:44, 10 November 2015 diff hist +828‎ deep Sparse Rectifier Neural Networks ‎ Add potential problems of rectified linear units.
• 12:5512:55, 10 November 2015 diff hist +150‎ deep Sparse Rectifier Neural Networks ‎ Add disadvantage of sparse representation.

4 November 2015

• 17:2817:28, 4 November 2015 diff hist +5,104‎ N neural Turing Machines ‎ Created page with "= Neural Turing Machines = Even though recurrent neural networks (RNNs) are [https://en.wikipedia.org/wiki/Turing_completeness Turing complete] in theory, the control of logical..."
• 17:2117:21, 4 November 2015 diff hist +35‎ m f15Stat946PaperSignUp ‎ Add link to paper summary.

2 November 2015

• 17:5017:50, 2 November 2015 diff hist +385‎ learning Hierarchical Features for Scene Labeling ‎ Add conclusions
• 17:4017:40, 2 November 2015 diff hist +97‎ learning Hierarchical Features for Scene Labeling ‎No edit summary
• 17:3617:36, 2 November 2015 diff hist +516‎ learning Hierarchical Features for Scene Labeling ‎ Add the big picture to introduction
• 17:2317:23, 2 November 2015 diff hist +368‎ learning Hierarchical Features for Scene Labeling ‎ Explain Laplacian

27 October 2015

• 13:4613:46, 27 October 2015 diff hist −37‎ m goingDeeperWithConvolutions ‎ Fix CNN image.
• 13:4613:46, 27 October 2015 diff hist 0‎ N File:Conv 1D nn.png ‎No edit summary current
• 13:4513:45, 27 October 2015 diff hist +1,494‎ goingDeeperWithConvolutions ‎ Adding some background on CNNs
• 13:2213:22, 27 October 2015 diff hist +735‎ goingDeeperWithConvolutions ‎ Elaborating on the inception module
• 13:0413:04, 27 October 2015 diff hist +207‎ goingDeeperWithConvolutions ‎ Adding to explanation of sparce matrix inefficiency.

23 October 2015

• 17:3917:39, 23 October 2015 diff hist −12‎ f15Stat946PaperSignUp ‎ Fix layout destroyed in previous change.
• 17:3717:37, 23 October 2015 diff hist +88‎ f15Stat946PaperSignUp ‎ Sign up for Set B paper

21 October 2015

• 14:5214:52, 21 October 2015 diff hist +1,090‎ stat946f15/Sequence to sequence learning with neural networks ‎ Add open questions
• 14:4014:40, 21 October 2015 diff hist +679‎ stat946f15/Sequence to sequence learning with neural networks ‎ More detailed discussion about the reversal of the input sequence.
• 14:1514:15, 21 October 2015 diff hist +1‎ m stat946f15/Sequence to sequence learning with neural networks ‎ typo
• 13:2513:25, 21 October 2015 diff hist +792‎ parsing natural scenes and natural language with recursive neural networks ‎ Adding summary of NLP results.
• 13:1513:15, 21 October 2015 diff hist +21‎ parsing natural scenes and natural language with recursive neural networks ‎ Add word representation dimensionality.
• 13:1413:14, 21 October 2015 diff hist −22‎ parsing natural scenes and natural language with recursive neural networks ‎ The paper does not state that these are the best values (it might just be values that work well enough).
• 13:1013:10, 21 October 2015 diff hist +209‎ parsing natural scenes and natural language with recursive neural networks ‎ Fix loss equation and clarify it.
• 12:5512:55, 21 October 2015 diff hist −1‎ m parsing natural scenes and natural language with recursive neural networks ‎ grammatics
• 12:5512:55, 21 October 2015 diff hist +1‎ m parsing natural scenes and natural language with recursive neural networks ‎ Replaced wrong word
• 12:5312:53, 21 October 2015 diff hist +14‎ m parsing natural scenes and natural language with recursive neural networks ‎ Fix lay
• 12:5012:50, 21 October 2015 diff hist +93‎ parsing natural scenes and natural language with recursive neural networks ‎ Clarifying word representation which is not a bag-of-words

5 October 2015

• 12:2212:22, 5 October 2015 diff hist +198‎ f15Stat946PaperSignUp ‎ Sign up for paper presentation