User contributions for Alcateri

17 December 2015

• 20:5720:57, 17 December 2015 diff hist +439‎ proposal for STAT946 (Deep Learning) final projects Fall 2015 ‎No edit summary

14 December 2015

• 13:4713:47, 14 December 2015 diff hist +623‎ generating text with recurrent neural networks ‎ →‎Discussion: Cleaned up the discussion and added a new idea about combining character and word level models.
• 13:3413:34, 14 December 2015 diff hist +1‎ m generating text with recurrent neural networks ‎ →‎Introduction
• 13:1213:12, 14 December 2015 diff hist +145‎ m neural Turing Machines ‎ →‎Neural Turing Machines: Cleaned up referencing and some wording
• 13:0813:08, 14 December 2015 diff hist +189‎ neural Turing Machines ‎ Added references section and cleaned up referencing in second paragraph

13 December 2015

• 00:1100:11, 13 December 2015 diff hist +339‎ m learning Phrase Representations ‎ →‎Experiments: - Added some specifics to the experimental results

12 December 2015

• 22:3122:31, 12 December 2015 diff hist +1,656‎ extracting and Composing Robust Features with Denoising Autoencoders ‎ →‎Analysis of the Denoising Autoencoder: - Included section on Stochastic Operator Perspective
• 21:5021:50, 12 December 2015 diff hist +35‎ m extracting and Composing Robust Features with Denoising Autoencoders ‎ →‎The Denoising Autoencoder: - Cleaned up the wording and mathematical notation in this section
• 19:0519:05, 12 December 2015 diff hist +2,040‎ the loss surfaces of multilayer networks (Choromanska et al.) ‎ Adding a "Prior Work" section, along with showing the references.

2 December 2015

• 15:1315:13, 2 December 2015 diff hist +564‎ on the difficulty of training recurrent neural networks ‎ →‎Background
• 15:0115:01, 2 December 2015 diff hist +40‎ on the difficulty of training recurrent neural networks ‎ →‎Background

23 November 2015

• 12:4012:40, 23 November 2015 diff hist +922‎ learning Fast Approximations of Sparse Coding ‎ →‎Coordinate Descent: Added CoD algorithm and comment about similarity to ISTA

19 November 2015

• 23:2123:21, 19 November 2015 diff hist +2,310‎ show, Attend and Tell: Neural Image Caption Generation with Visual Attention ‎ Adding Related Work section

18 November 2015

• 20:4320:43, 18 November 2015 diff hist +653‎ the Manifold Tangent Classifier ‎ →‎Discussion
• 18:1618:16, 18 November 2015 diff hist +20,731‎ N the Manifold Tangent Classifier ‎ Created page with "== Introduction == The goal in many machine learning problems is to extract information from data with minimal prior knowledge<ref name = "main"> Rifai, S., Dauphin, Y. N., Vinc..."
• 17:1617:16, 18 November 2015 diff hist +44‎ N File:Figure 1 MTC.png ‎ Visualization of the learned tangents of CAE current

16 November 2015

• 12:3912:39, 16 November 2015 diff hist +44‎ f15Stat946PaperSignUp ‎ →‎Set B

12 November 2015

• 11:5411:54, 12 November 2015 diff hist +2,673‎ deep neural networks for acoustic modeling in speech recognition ‎ →‎Generative Pretraining
• 09:2509:25, 12 November 2015 diff hist +622‎ deep neural networks for acoustic modeling in speech recognition ‎ →‎Generative Pretraining: - Introduction of Generative Pretraining

11 November 2015

• 22:2422:24, 11 November 2015 diff hist +31‎ deep neural networks for acoustic modeling in speech recognition ‎No edit summary

30 October 2015

• 10:1510:15, 30 October 2015 diff hist −2‎ human-level control through deep reinforcement learning ‎ →‎The Bellman Equation in the Loss Framework
• 10:1010:10, 30 October 2015 diff hist +3‎ human-level control through deep reinforcement learning ‎ →‎Data & Preprocessing
• 10:0510:05, 30 October 2015 diff hist −34‎ human-level control through deep reinforcement learning ‎ →‎Introduction
• 10:0510:05, 30 October 2015 diff hist −75‎ human-level control through deep reinforcement learning ‎ →‎Introduction
• 10:0410:04, 30 October 2015 diff hist −10‎ human-level control through deep reinforcement learning ‎ →‎Introduction
• 10:0310:03, 30 October 2015 diff hist +769‎ human-level control through deep reinforcement learning ‎No edit summary
• 09:5009:50, 30 October 2015 diff hist −8‎ human-level control through deep reinforcement learning ‎ →‎Results
• 09:5009:50, 30 October 2015 diff hist +238‎ human-level control through deep reinforcement learning ‎ →‎Results with model components removed
• 07:5207:52, 30 October 2015 diff hist +65‎ human-level control through deep reinforcement learning ‎ →‎Results with model components removed
• 07:4907:49, 30 October 2015 diff hist −4‎ human-level control through deep reinforcement learning ‎ →‎Results with model components removed
• 07:4807:48, 30 October 2015 diff hist +737‎ human-level control through deep reinforcement learning ‎No edit summary

29 October 2015

• 23:0423:04, 29 October 2015 diff hist +212‎ human-level control through deep reinforcement learning ‎No edit summary
• 23:0023:00, 29 October 2015 diff hist +56‎ N File:Performance.JPG ‎ Performance of DQN vs. Human Tester and Existing Results current
• 22:5922:59, 29 October 2015 diff hist +873‎ human-level control through deep reinforcement learning ‎No edit summary
• 22:4722:47, 29 October 2015 diff hist +359‎ human-level control through deep reinforcement learning ‎No edit summary
• 22:1122:11, 29 October 2015 diff hist +56‎ N File:QLearning Alg.JPG ‎ The algorithm for deep Q-learning with Experience Replay current
• 22:1022:10, 29 October 2015 diff hist +1,529‎ human-level control through deep reinforcement learning ‎No edit summary
• 21:3621:36, 29 October 2015 diff hist +682‎ human-level control through deep reinforcement learning ‎ →‎Training
• 20:5120:51, 29 October 2015 diff hist +156‎ human-level control through deep reinforcement learning ‎No edit summary
• 20:4920:49, 29 October 2015 diff hist +2‎ human-level control through deep reinforcement learning ‎ →‎Training Details
• 20:4820:48, 29 October 2015 diff hist −3‎ human-level control through deep reinforcement learning ‎ →‎Training Details
• 20:4720:47, 29 October 2015 diff hist −13‎ human-level control through deep reinforcement learning ‎ →‎Training Details
• 20:3120:31, 29 October 2015 diff hist −27‎ human-level control through deep reinforcement learning ‎No edit summary
• 20:2920:29, 29 October 2015 diff hist +1,540‎ human-level control through deep reinforcement learning ‎No edit summary
• 20:0220:02, 29 October 2015 diff hist +66‎ N File:Network Architecture.JPG ‎ A visualization of the deep convolutional network that estimates Q current
• 18:2218:22, 29 October 2015 diff hist +902‎ human-level control through deep reinforcement learning ‎No edit summary
• 17:4217:42, 29 October 2015 diff hist +1‎ human-level control through deep reinforcement learning ‎No edit summary
• 17:4117:41, 29 October 2015 diff hist +1,559‎ human-level control through deep reinforcement learning ‎ →‎Methodology
• 11:3411:34, 29 October 2015 diff hist +5,637‎ N human-level control through deep reinforcement learning ‎ Created page with "== Introduction == Reinforcement learning is "the study of how animals and artificial systems can learn to optimize their behaviour in the face of rewards and punishments" <ref>..."

23 October 2015

• 11:1511:15, 23 October 2015 diff hist +47‎ parsing natural scenes and natural language with recursive neural networks ‎ →‎Recursive Neural Networks for Structure Prediction: Improved readability of math expressions by inserting "\,"