Differences

This shows you the differences between two versions of the page.

--- cs601r_w2020:proj1 [2020/01/22 17:49]
wingated
+++ cs601r_w2020:proj1 [2021/06/30 23:42]
@@ Line 1: / Line 1: @@
-====Objective:====
-To explore significant deep learning projects.
-----
-====Deliverable:====
-There are two deliverables for Projects 1 & 2:
-  * An excel spreadsheet (or CSV file) that shows the total amount of time you spent on your final, broken down by day
-  * A PDF writeup of your project (3-6 pages)
-----
-====Grading standards:====
-Each project counts for about 20-25% of your overall grade (see Learning Suite for a precise breakdown of the value of different assignments).
-Unlike the final project, grading of your projects will be done primarily based on your writeups.
-I will grade your writeups based on:
-  * The careful use of the scientific method in your work
-  * The quality of the writing
-I am expecting about 25 hours of effort on each project.
-----
-====Description:====
-For your projects, you should execute a substantial project of your own choosing.  You will turn in a single writeup (in PDF format only, please!).  Your writeup can be structured in whatever way makes sense for your project, but see below for some possible outlines.
-----
-====Requirements for the time log:====
-For the time log, you must document the time you spent (on a daily basis) along with a simple description of your activities during that time.  **If you do not document your time, it will not count.**  In other words, it is not acceptable to claim that you spent 35 hours on your project, without a time log to back it up.  I will not accept any excuses about this requirement.
-So, for example, a time log might look like the following:
-  * 8/11 - 1 hour - read alphago paper
-  * 8/12 - 2 hours - downloaded and cleaned data
-  * 8/21 - 4 hours - found alphago code
-  * 8/24 - 1 hour - implemented game logic
-  * 9/17 - 2 hours - worked on self-play engine
-  * 9/18 - 1 hour - worked on self-play engine
-  * 10/1 - 2 hours - started training
-  * ... etc.
-Additional requirements:
-  * You may not count any more than 5 hours of research and reading
-  * You may not count any more than 10 hours of "prep work".  This could include dataset preparation, collection and cleaning; or wrestling with getting a simulator / model working for a deep RL project; etc.
-  * At least 10 hours must involve designing, testing, and iterating deep learning-based models, analyzing results, experimenting, etc.
-  * You don't get extra credit for more than 25 hours.  Sorry.  :)
-----
-====Requirements for the writeup:====
-Your writeup serves to inform me about what you did, and simply needs to describe what you did for your project.
-Ideally, your writeup would look and feel like an academic paper in your discipline.
-The primary thing I will be looking for is the **systematic use of the scientific method** in your project.  That is, you should have a clearly stated problem, a clearly stated hypothesis, and experiments to test your ideas.
-It is unlikely that your first idea will pan out.  **A critical part of the writeup is that I should be able to see an iterative approach to refining your solution.**
-In addition, you should describe:
-  * The problem you set out to solve
-  * The exploratory data analysis you did
-  * Your technical approach
-  * Your results
-It should be about 3-6 pages.
-----
-====Possible project ideas:====
-Many different kinds of final projects are possible.  A few examples include:
-  * Learning how to render a scene based on examples of position and lighting
-  * Learning which way is "up" in a photo (useful for drone odometry)
-  * Training an HTTP server to predict which web pages a user will likely visit next
-  * Training an earthquake predictor
-  * Using GANs to turn rendered faces into something more realistic (avoiding the "uncanny valley")
-  * Transforming Minecraft into a more realistic looking game with DNN post-processing
-  * Using style transfer on a network trained for facial recognition (to identify and accentuate facial characteristics)
-  * Using RGB+Depth datasets to improve geometric plausibility of GANs
-The project can involve any application area, but the core challenge must be tackled using some sort of deep learning.
-The best projects involve a new, substantive idea and novel dataset.  It may also be acceptable to use vanilla DNN techniques on a novel dataset, as long as you demonstrate significant effort in the "science" of the project -- evaluating results, exploring topologies, thinking hard about how to train, and careful test/training evaluation.  It may also be acceptable to simply implement a state-of-the-art method from the literature, but clear such projects with me first.
-----
-====Notes:====
-You are welcome to use any publicly available code on the internet to help you.
-Here are some possible questions that you might consider answering as part of your report:
-  - **A discussion of the dataset**
-    - Where did it come from?  Who published it?
-    - Who cares about this data?
-  - **A discussion of the problem to be solved**
-    - Is this a classification problem?  A regression problem?
-    - Is it supervised?  Unsupervised?
-    - What sort of background knowledge do you have that you could bring to bear on this problem?
-    - What other approaches have been tried?  How did they fare?
-  - **A discussion of your exploration of the dataset**.
-    - Before you start coding, you should look at the data.  What does it include?  What patterns do you see?
-    - Any visualizations about the data you deem relevant
-  - **A clear, technical description of your approach.**
-    - Background on the approach
-    - Description of the model you use
-    - Description of the inference / training algorithm you use
-    - Description of how you partitioned your data into a test/training split
-    - How many parameters does your model have?  What optimizer did you use?
-    - What topology did you choose, and why?
-    - Did you use any pre-trained weights?  Where did they come from?
-  - **An analysis of how your approach worked on the dataset**
-    - What was your final RMSE on your private test/training split?
-    - Did you overfit?  How do you know?
-    - Was your first algorithm the one you ultimately used for your submission?  Why did you (or didn't you) iterate your design?
-    - Did you solve (or make any progress on) the problem you set out to solve?

BYU CS classes

User Tools

Site Tools

Differences

Page Tools