Final Homework
- 5/17 - 31 Reading presentation (All details)
- 5/17 : A Tutorial on Deep Learning, Quoc V. Le (Research Scientist, Google), 2015/10.
- 5/24 : Neural Networks and Deep Learning, Michael Nielsen, 2015.
- Ch 1: Using neural nets to recognize handwritten digits. Presenter: Jacky.
- Ch 2: How the backpropagation algorithm works, Presenter: Cindy.
- Ch 3: Improving the way neural networks learn, Presenters: Karissa, Dioxin.
- 5/31 : Neural Networks and Deep Learning, Michael Nielsen, 2015.
- Ch 4: A visual proof that neural nets can compute any function Presenter: Kirito.
- Ch 5: Why are deep neural networks hard to train? Presenter: Daniel.
- Ch 6: Deep learning Presenters: Tommy, Lolly.
- 6/7 : Group discussion
- Two discussion groups
- Neural Networks group: Jonathan, spencer, jacky, cindy, karissa, dioxin. Leader: Jonathan. Scouts:Jacky, Dioxin.
- Deep Neural Networks group: Steffi, kirito, daniel, tommy, lolly. Leader: Steffi. Scouts:Kirito,Lolly,
- Two question sheets: NN questions. DNN questions.
- The teacher will give the two groups with two question sheets: one for NN, and another for DNN.
- Each group has to complete one question sheet with intra-group discussion, but complete another sheet with inter-group discussion.
- Roles in a group: leader, scout and member. Each group has one leader and two scouts.
- Leader is responsible to lead the intra-group and inter-group discussion.
- Scouts are responsible to join inter-group discussion and complete another sheet.
- Plan of schedule: 1.5 hour for intra-group discussion, 1 hour for inter-group discussion.
- 0 - 1 hr: Intra-group discussion lead by the leader.
- 1 - 1.5 hr: Scouts go out to another group to join inter-group discussion, which is still lead by the leader.
- 1.5 - 2.5 hr: Scouts come back to orignial group, and explain the answer of another question sheet to group members.
- Final result :
- Every person should complete the two sheets, and upload the two sheets (ex. modify the docx file or take photos) to the course web site.
- Two discussion groups
- 6/14 Programming presentation
- Requirements
- Your programming topic should relate to your presentation.
- Every student has 10 minutes to present and demo your programming results.
- Upload your programming presentation (pptx/pdf) before 24:00, 6/13.
- Click this link to upload your file.
- Name your file with: month_day_name. Ex.: 06_14_Jacky.pptx.
- Presentation should include:
- (1) Explanation of the theory of the program.
- (2) Input and output examples of the program.
- (3) Live-show demo of the program.
- Suggested topics of programming
- Matlab neural network toolbox examples
- Python examples in : Neural Networks and Deep Learning, Michael Nielsen, 2015.
- Study of deep neural network toolkit
- Caffe, TensorFlow, ConvNetJS, Deep Learning with Matlab by Mathworks, and so on.
- Demonstration programs of deep neural network papers.
- 6/21 Final Report
- Final reports: Steffi, Jonathan, Daniel, Jacky. Lolly, Cindy, Tommy. Kirito. Karissa. Dioxin. Spencer.
- Requirements
- The length of this final report is required to be at least 3000 words with English.
- Plagiarism is not allowed.
- Upload your final report (Weebly web page) to your web site before 6/21 24:00.
- Suggestions
- Content
- You can write everything you learned in this course.
- But you can focus more on what you learned in this final homework.
- Structure
- Introduction (what is PR, what is NN/DNN, why NN/DNN can be a good PR method)
- Your reading contents
- Your learning from group discussion
- Your programming contents
- Conclusion
- Content