CS 519, Applied Machine Learning, Spring 2022, Paper Review

Requirements

non-technical background (3 pts):

• Who: Who are the authors? Which institutions are they from? Who are the PI and who are the students? Are there any interns?
• Where: Where and when was this paper published? Note: it's possible that a paper is first published in a conference and later expanded to a journal paper, so check if this is the case. Did this paper receive any recognition such as an award or nomination? Does this paper have impact (number of citations)? Is the impact increasing or decreasing (trend in the number of citations)? Are there media reports about this paper?
• Reproducibility: Did the authors create and release any dataset? If not, what the datasets did they use, and are they freely available online? Is there a clear train/dev/test split? Did the authors release their code? Are these datasets and code influential? Is there a demo? Can you find slides or talk videos about this paper? Overall, how reproducible is this paper?

  core (the famous what-why-how-wow template for writing abstract/intro) (8 pts):

• What: What problem are the authors trying to solve? Is this a new problem? If not, what are the notable previous approaches, and what are their limitations?
• Why: Why did they choose this problem? Is it hard? Is it important?
• How: What's the authors' approach to tackle this problem?
• Wow: Did this approach result in great results?

  further: (3 pts)

• But: What flaws or limitations did you see in the authors' approach or results?
• More: If you were to follow up on this line of work, what topics/methods will you work on?
• All: What's your overall feeling of this paper? What's the single take-home message you learned from it?

  relevance: (1 pt)

• Did materials covered in this course help you understand this paper? How relevant is this paper to this course?
• What extra materials did you study in order to understand this paper?

List of Candidate Papers (you must choose a paper from this list)

bias, bias amplification, and fairness in machine learning

1. Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints (2017)
2. Feature-Wise Bias Amplification (2018)
3. Equality of Opportunity in Supervised Learning (2016)

   large-scale machine learning

4. XGBoost

   question answering and truly-blind competitions

5. SQuAD 1.0 (2016)
6. SQuAD 2.0 (2018)
7. CoQA (2018)

   deep reinforcement learning

8. Atari, Nature, 2015
9. AlphaGo, Nature 2016
10. AlphaGoZero, Nature 2017
11. AlphaZero, Science 2018

    deep learning for language and vision

12. show and tell: Show and Tell: A Neural Image Caption Generator, 2014--2015
13. show, attend, and tell: Show, Attend, and Tell: Neural Image Caption Generation with Visual Attention, 2015--2016
14. neural machine translation: Neural Machine Translation by Jointly Learning to Align and Translate ICLR 2015
15. sentence representation: Dependency-based Convolutional Neural Networks for Sentence Embeddings, ACL 2015
16. listen and tell: Audio Caption: Listen and Tell, ICASSP 2019
17. Transformer: Attention Is All You Need, NIPS 2017
18. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, EMNLP 2018

    deep learning for biology

19. protein folding: Distance-based protein folding powered by deep learning, PNAS 2019
20. protein folding: AlphaFold: Improved protein structure prediction using potentials from deep learning, Nature 2020
21. protein folding: AlphaFold2: Highly accurate protein structure prediction with AlphaFold, Nature 2021
22. RNA folding: RNA secondary structure prediction using deep learning with thermodynamic integration, Nature Communications 2021
23. BERT on protein sequences: Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences, PNAS 2021

    deep learning for programming languages

24. Tree-to-tree Neural Networks for Program Translation, NIPS 2018
25. AlphaCode: Competition-Level Code Generation with AlphaCode, preprint 2022