Project Options

There are two project options. All require that you submit a written proposal by March. 21st online.

1. Propose your own topic.
2. Pick a recent paper from a top-tier architecture conference and validate part of its results. You may choose to validate one of the papers that were discussed in the course. I will provide suggestion. Preference will be for accelerators and more so for Machine Learning accelerators.

I will set aside a day for meeting in person and discussing your plans prior to submitting your proposal. My current plan is to hold the meetings on March 14th or 15th.

You will have to deliver the following:

1. Proposal, due March 21st.
2. A tar.gz file with all source files plus how to compile. Due April 30th.
3. A final report. Due April 30th.
4. A presentation. Due April 30th.

See below for format/suggestions.

This should be an at most two pages. One page is just fine – don't feel that you have to write lots. You should explain in the following order:

1. Topic, i.e., what are you going to do. Start by motivating the particular work. Give a set of reasons why this is relevant (not from an educational perspective but from a computer design perspective).

2. Methodology and Goals. For CPU related projects you will be expected to use Simplescalar and the SPEC CPU 2000 benchmarks that will be provided. For ML acceleration we will provide you with some models and traces and skeleton code to use them. In your proposal you should explain which experiments you will be performing and why. Try to justify why these experiments are the most relevant given the amount of time you have available. This will be our contract: this is what you promise to do. If you are validating an existing study, say which parts you will validate. You are not expected to go over all results and validate everything. A small subset of important results is good enough. The most important part is picking which results are important. You may also want to discuss other results not found in the paper that you think complement the work and you are willing to try to get.

Remember: while these may evolve or change you are required to start with a meaningful plan. No point starting something if you cannot articulate why this might be interesting or doable.

Try to limit this to at most five pages. (Remember: it’s easy to write a lot of text. It’s hard to write concise.) An approximate format is:

1. Introduction: Motivation and Problem statement. BRIEFLY. Also conclude with forecasting what your method and most important results are.

2. Expanded motivation if needed.

3. Related Work. Please emphasize this section if you decide to validate a set of published results. The publications suggested later in this document are not all very recent. It is your responsibility to find and report any related work.

4. If you are proposing a new mechanism describe it.

5. Methodology. Which benchmarks, what simulation method and which simulator, machine configuration and any other relevant parameters. Also a brief summary of your metrics and WHY are you using them

6. Evaluation. One by one the results. Explain the results. Why do they look they way they do? Are they correct? Can you come up with examples that show that the trends exhibited are possible?

7. Conclusions. “I am so good I cannot stand myself. Maybe I should donate my brain to science” put in less obvious terms. Future directions (not left or right).

If for whatever reason you are not able to deliver exactly what your promised, please explain why. Please keep this in mind: I'm more interested to see how you think. It's better to do less and do it well than to do lots and do them poorly.

You will be required to give a presentation at the end of the course. We will schedule these. They will be 10' to 15'.

• Bit-Fusion
• Efficient Inference Engine
• SnapPea
• SCNN
• Cambricon-X
• ShiDaDianNao
• Fused-layer CNN accelerators
• TETRIS
• ESE: efficient speech recognition engine with sparse LSTM on FPGA
• Or take a look at this list https://github.com/fengbintu/Neural-Networks-on-Silicon

And feel free to search online for others. There is tons of very interesting and excellent work that I have not had the chance to include above. By no means is the above a complete list. I listed works that I thought would be “reasonable” to try to recreate (in part) as part of a course project. Keep in mind, we are validating work. We get the benefit of knowing what to look for and what to evaluate. It is a lot harder and time consuming to start from scratch. There was a lot more time/work spent to arrive at these configurations and choices.