Artificiality Bites 💊 Issue #45
Hello Human! This is a new issue from my weekly newsletter, holding a tiny compilation made of interesting articles from last week, projects, tutorials and tools; all related to Data, Artificial Intelligence and adjacent topics. Njóttu máltíðarinnar!
📝 Interesting publications this week
- Pytorch-Widedeep, deep learning for tabular data IV: Deep Learning vs LightGBM
56'
A thorough comparison between Deep Learning algorithms and LightGBM for tabular data for classification and regression problems. - Teacher Algorithms for Deep RL Agents that Generalize in Procedurally Generated Environments
19'
Teaching algorithms are becoming a key ingredient to scaffold Deep Reinforcement Learning agents into their learning journey, leading to policies that can generalize to multiple environments. This publication presents this emerging sub-field and showcases recent work done in the Flowers Lab on designing a Learning Progress based teacher algorithm. - AI can now emulate text style in images in one shot using just a single word
8'
Facebook AI announced TextStyleBrush, an AI research project that can copy the style of text in a photo using just a single word. With this AI model, you can edit and replace text in images.
- The Beginner’s Guide to the Modern Data Stack
4'
A curated list of blogs, books, newsletters, podcasts, and communities for all things modern data stack. - 10 steps to educate your company on AI fairness
6'
As part of the World Economic Forum's Global Future Council on AI for Humanity, a collective of AI practitioners, researchers and corporate advisors, proposed 10 practical interventions for companies to employ in order to ensure AI fairness. - The Rise of the Metadata Lake
7'
Introducing a new way of storing metadata for today’s limitless use cases like data discovery, lineage, observability and fabrics.
🔧 Tutorials
- GPT-J-6B Inference Demo
This notebook demonstrates how to run the GPT-J-6B model. See the link for more details about the model, including evaluation metrics and credits. - Facial Landmarks, a Solution in Deepfakes
9'
There are several different methods we can use to detect facial landmarks as features for the task of fake content generation. This publication covers three of the most widely used methods: OpenCV, dlib, and MTCNN. - Image Captioning with Keras
13'
Implement an image captioning model using a CNN and a Transformer.
📦 Repositories
- speechbrain/speechbrain
SpeechBrain is a toolkit for developing state-of-the-art speech systems, like speech recognition, speaker recognition, speech enhancement, multi-microphone signal processing and many others. - facebookresearch/AugLy
Augly is a data augmentation library for audio, image, text and video, supporting over 100 different augmentations.
- PrithivirajDamodaran/Styleformer
A Neural Language Style Transfer framework to transfer natural language text smoothly between fine-grained language styles like formal/casual, active/passive, and many more. - serengil/chefboost
Chefboost is a lightweight decision tree framework for Python with categorical feature support. It covers regular decision tree algorithms (ID3, C4.5, CART, CHAID and regression tree) plus some advanced techniques (gradient boosting, random forest and adaboost). - shankarpandala/lazypredict
Lazy Predict builds a lot of basic models with a few lines of code in order to figure out which models work better without any parameter tuning. - neuralchen/SimSwap
An Efficient Framework For High Fidelity Face Swapping.
- castor-team/airflow-castor
A framework for building Airflow DAGs via YAML files.
🎓 Courses / Books / Events
- Reproducible Data Science 📕
Accessible data analysis with open source Python tools and real-world data by Valentin Danchev. - The Principles of Deep Learning Theory 📘
449p
An Effective Theory Approach to Understanding Neural Networks, written by Daniel A. Roberts and Sho Yaida.
🚀 Extra bits
👋 See you next week!