Interesting 2021 paper roundup from a CMU econ prof.

Lots on forecasting!

https://donskerclass.github.io/post/papers-i-liked-2021


Team
@data_enthusiasts
@OpenArchiveBooks

Practical image restoration

Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data

Research Paper:
https://arxiv.org/pdf/2107.10833v1.pdf

Code:
https://github.com/xinntao/Real-ESRGAN

Colab demo:
https://colab.research.google.com/drive/1k2Zod6kSHEvraybHl50Lys0LerhyTMCo?usp=sharing

Team
@data_enthusiasts
@OpenArchiveBooks

From Google and Waymo researchers: The self-/unsupervised revolution is near! Unsupervised optical flow model SMURF improves SOTA by 40% and beats many supervised methods such as PWC-Net and FlowNet2

SMURF: Self-Teaching Multi-Frame Unsupervised RAFT with Full-Image Warping

Research Paper:
https://arxiv.org/pdf/2105.07014.pdf

Video:
https://www.youtube.com/watch?v=W7NCbfZp6QE

Code:
https://github.com/google-research/google-research/tree/master/smurf

Team
@data_enthusiasts
@OpenArchiveBooks

A Beginner's Guide to Clean Data (online book)

https://b-greve.gitbook.io/beginners-guide-to-clean-data/

Team
@data_enthusiasts
@OpenArchiveBooks

Balanced Chamfer Distance as a Comprehensive Metric for Point Cloud Completion

Github: https://github.com/wutong16/density_aware_chamfer_distance

Paper: https://arxiv.org/pdf/2111.12702

Dataset: https://paperswithcode.com/dataset/mvp


Team
@data_enthusiasts
@OpenArchiveBooks

A quick review of GPT-3 | What is it and how does it work?

https://www.youtube.com/watch?v=xB6hZwYsV2c

Team
@data_enthusiasts
@OpenArchiveBooks

Learning perturbation sets for robust machine learning

Tackled the problem of defining a perturbation set for real-world perturbations which cannot be easily described with a set of equations.

Paper: https://arxiv.org/pdf/2007.08450

Blog post: https://locuslab.github.io/2020-07-20-perturbation/

Code: https://github.com/locuslab/perturbation_learning

Team
@data_enthusiasts
@OpenArchiveBooks

Here is a new, and free book on Machine Learning from scratch.

It includes the math and code examples. Solid reference.

https://dafriedman97.github.io/mlbook/content/table_of_contents.html

Team
@data_enthusiasts
@OpenArchiveBooks

🗣 AV-HuBERT (Audio-Visual Hidden Unit BERT)

AV-HuBERT is a self-supervised representation learning framework for audio-visual speech.

Code: https://github.com/facebookresearch/av_hubert

Facebook AI: https://ai.facebook.com/blog/ai-that-understands-speech-by-looking-as-well-as-hearing/

Paper: https://arxiv.org/pdf/2201.02184

Team
@data_enthusiasts
@OpenArchiveBooks

6 Best Artificial Intelligence Courses for Healthcare in 2022

https://www.mltut.com/best-artificial-intelligence-courses-for-healthcare/

Team
@data_enthusiasts
@OpenArchiveBooks

PyRetri: An open source deep learning based unsupervised image retrieval toolbox built on PyTorch🔥

Content-based Image retrieval

Paper: https://arxiv.org/pdf/2005.02154

Code: https://github.com/PyRetri/PyRetri

Installation: https://github.com/PyRetri/PyRetri/blob/master/docs/INSTALL.md

Team
@data_enthusiasts
@OpenArchiveBooks

Using public datasets in commercial software

Great paper: Can I use this publicly available dataset to build commercial AI software? Most likely not.

An extremely important question for the Data Science community: it turns out (wat?) that not all publicly available datasets can be used to build commercial solutions 💣💣💣.

Authors examine license agreements of 6 popular datasets used in Computer Vision (CIFAR-10, ImageNet, Cityscapes, FFHQ, VGGFaces2 and MS COCO) and conclude that the models trained on these data can not be commercialized at least.

An example of the results of the CIFAR-10 dataset's license analysis is shown in the screenshot.

It was logical to assume this, but for many community members this may be the opening of the century.

Team
@data_enthusiasts
@OpenArchiveBooks

Deep Learning Interviews:

Hundreds of fully solved job interview questions from a wide range of key topics in AI

https://arxiv.org/pdf/2201.00650

Team
@data_enthusiasts
@OpenArchiveBooks

Machine Learning for Everyone with great explanation

https://vas3k.com/blog/machine_learning/?fbclid=IwAR3e3_Yy7LyEm7-Yqvuttx_zftCn5hL8NXBRSaut0gp7sT8AlxDbj-q3dNQ

and find other free-resources:

https://www.marktechpost.com/free-resources/?fbclid=IwAR2XJOTWGEdC9aFSRibXWZAZlLpJY9nmRnS_PDfNOgllc9o_mL3OxJaQleY

Team
@data_enthusiasts
@OpenArchiveBooks

A love story involving choice versus predestination.

Feedback is encouraged.

"humAIn" - a short story

https://carpevelo.blogspot.com/2021/07/humain-short-story.html?m=0

https://redd.it/s5tcif

Team
@data_enthusiasts
@OpenArchiveBooks

Learn to Build Machine Learning Models

Hey, I’m Merve from Hugging Face, an Open-Source company working in the democratization of responsible Machine Learning. 👋

We want to reduce the barrier of entry for people that want to learn about machine learning and create awesome products with ML. Our recent Open-Source initiative is called “Tasks”, in which we've curated use cases, tutorials, models, datasets and more to help you in your path in building models. 🙌🏼

We appreciate feedbacks 🤗 Let us know what you think!

You can get started with learning here. You can also use hf.co/tasks shortcut. 🤓


Team
@data_enthusiasts
@OpenArchiveBooks

💡 Introducing StylEx: A New Approach for Visual Explanation of Classifiers

Github: https://explaining-in-style.github.io/

Code: https://github.com/google/explaining-in-style

Article: https://ai.googleblog.com/2022/01/introducing-stylex-new-approach-for.html

Video: https://explaining-in-style.github.io/#video

Team
@data_enthusiasts
@OpenArchiveBooks

Beginner's guide to Python for Data Science by atoti

Team
@data_enthusiasts
@OpenArchiveBooks

🔥 The first high-performance self-supervised algorithm that works for speech, vision, and text

Github: https://github.com/pytorch/fairseq/tree/main/examples/data2vec

Paper: https://ai.facebook.com/research/data2vec-a-general-framework-for-self-supervised-learning-in-speech-vision-and-language

Meta AI: https://ai.facebook.com/blog/the-first-high-performance-self-supervised-algorithm-that-works-for-speech-vision-and-text

Team
@data_enthusiasts
@OpenArchiveBooks

ConvMixer, an extremely simple model that is similar in spirit to the ViT and the even-more-basic MLP-Mixer

Github: https://github.com/locuslab/convmixer

Paper: https://arxiv.org/pdf/2201.09792v1.pdf

Team
@data_enthusiasts
@OpenArchiveBooks