[Title of paper]

1

Presenter: [Name]

[Date]

CS 6758: Deep Learning for Robotics

Motivation and Main Problem

1-5 slides

High-level description of problem being solved

Why is the problem important?

• its significance towards general-purpose robot autonomy
• its potential application and societal impact of the problem

Technical challenges arising from the problem

• the role of the AI and machine learning in tackling this problem

High-level idea of why prior approaches didn’t already solve

Key insight(s) (try to do in 1-3) of the proposed work

​

2

CS 6758: Deep Learning for Robotics

Problem Setting

1 or more slides

Problem formulation, key definitions and notations

• Be precise -- should be as formal as in the paper

3

CS 6758: Deep Learning for Robotics

Context / Related Work / Limitations of Prior Work

1 or more slides

Which other papers have tried to tackle this problem or a related problem?

• The paper’s related work is a good start, but there may be others
• What is the key limitations of prior work(s)?

​

4

CS 6758: Deep Learning for Robotics

Proposed Approach / Algorithm / Method

5

1-5 slides

Describe algorithm or framework (pseudocode and flowcharts can help)

• What is the optimization objective?
• What are the core technical innovations of the algorithm/framework?

​

Implementation details should be left out here, but may be discussed later if its relevant for limitations / experiments

CS 6758: Deep Learning for Robotics

Theory (if relevant)

6

What are the assumptions made for the theory? Are these reasonable? Realistic?

​

If the theory build strongly on other prior theory / results, reference those and state them here.

​

CS 6758: Deep Learning for Robotics

Theory (if relevant, continued)

7

State main results formally

Give proof sketches

Refer students to the full proofs in paper

CS 6758: Deep Learning for Robotics

Experimental Setup

8

1-3 slides

Description of the experimental evaluation setting

• What is the domain(s), e.g., datasets, tasks, robot hardware setups?
• What are the baseline(s)?
• What scientific hypotheses are tested?

​

How did the authors evaluate the success of their approach?

• Clear description of the metrics that will be used

CS 6758: Deep Learning for Robotics

Experimental Results

9

>1 slide

Present the quantitative and qualitative results

Show figures / tables / plots / robot demos

Pinpoint the most interesting / significant results

CS 6758: Deep Learning for Robotics

Discussion of Results

10

1-2 slides

What conclusions are drawn from the results by the authors?

• What insights are gained from the experiments?
• What strengths and weaknesses of the proposed method are illustrated by the results?

​

Are the stated conclusions fully backed by the results and references?

• If so, why? (Recap the relevant supporting evidences from the given results + refs)
• If not, what are the additional experiments / comparisons that can further support/repudiate the conclusions of the paper?

CS 6758: Deep Learning for Robotics

Critique / Limitations / Open Issues

11

1-2 slides

What are the key limitations of the proposed approach / ideas? (e.g. does it require strong assumptions that are unlikely to be practical? Computationally expensive? Require a lot of data?)

​

Are there any practical challenges in deploying the approach on physical robots in the real world? Are there any safety or ethical concerns of using such approach?

​

If follow-up work has addressed some of these limitations, include pointers to that. But don’t limit your discussion only to the problems / limitations that have already been addressed.

CS 6758: Deep Learning for Robotics

Future Work for Paper / Reading

12

1-2 slides

What interesting questions does it raise for future work?

• Your own ideas for future work
• Others’ ideas (if others have already built on this idea)

CS 6758: Deep Learning for Robotics

Extended Readings

13

1-2 slides

Pointers to papers that use this paper as a reference and/or other very related papers that others may want to read

​

Point classmates to where they can go for further reading on this paper/reading

CS 6758: Deep Learning for Robotics

Summary

14

1 slide

Approximately one bullet for each of the following

​

• Problem the reading is discussing
• Why is it important and hard
• What is the key limitation of prior work
• What is the key insight(s) (try to do in 1-3) of the proposed work
• What did they demonstrate by this insight? (tighter theoretical bounds, state of the art performance on X, etc)

CS 6758: Deep Learning for Robotics