True Positive Weekly #35

The most important artificial intelligence and machine learning links of the week

Hey, in this issue: Facebook developed a "neural database"; paper explained: Fastformer; Tutorial on inferring concept drift without labeled data; a comparison between the investment returns of business angels and machine learning algorithms; a tutorial on human pose classification; and more.

The sponsor of this issue is InterviewQs.

• Facebook developed a "neural database" that enables machines to search unstructured data (paper 1, paper 2, code)

• [Paper explained] Fastformer: Additive Attention Can Be All You Need

• [Tutorial] Inferring concept drift without labeled data

• How big data carried graph theory into new dimensions

• [Sponsored] InterviewQs - Get better at data science interviews by solving a few questions per week for free

• A comparison between the investment returns of business angels and machine learning algorithms

• Text data augmentation for deep learning

• Hora: an approximate nearest neighbor search algorithm written in Rust with a Python API

• [Tutorial] Human pose classification with MoveNet and TensorFlow Lite

Enjoy the newsletter? Help us make it bigger and better by sharing it with your colleagues and friends.

Have a nice weekend! See you next week. — Andriy