Every two weeks, we find the most interesting data science links from around the web and collect them in Data Science Briefings, the DataMiningApps newsletter. Subscribe now for free if you want to be the first to get up to speed on interesting resources.
- Flawed Algorithms Are Grading Millions of Students’ Essays
Fooled by gibberish and highly susceptible to human bias, automated essay-scoring systems are being increasingly adopted, a Motherboard investigation has found.
- The risks of amoral AI
The consequences of deploying automation without considering ethics could be disastrous
- Exploring Weight Agnostic Neural Networks
“We present a first step toward searching specifically for networks with these biases: neural net architectures that can already perform various tasks, even when they use a random shared weight.”
- Natural Language Processing: the age of Transformers
This article is the first installment of a two-post series on Building a machine reading comprehension system using the latest advances in deep learning for NLP. Very readable and thorough!
- A 2019 Guide to Speech Synthesis with Deep Learning
Artificial production of human speech is known as speech synthesis. This machine learning-based technique is applicable in text-to-speech, music generation, speech generation, speech-enabled devices, navigation systems, and accessibility for visually-impaired people.
- The State of Transfer Learning in NLP
“In the span of little more than a year, transfer learning in the form of pretrained language models has become ubiquitous in NLP and has contributed to the state of the art on a wide range of tasks.”
- Neural networks (NNs) are in essence polynomial regression (PR)
A provocative paper. Our advise is to read the results with a grain of salt: the data sets and neural networks used are pretty tiny.
- New State of the Art AI Optimizer: Rectified Adam (RAdam). Improve your AI accuracy instantly versus Adam, and why it works
“A new paper by Liu, Jian, He et al introduces RAdam, or “Rectified Adam”. It’s a new variation of the classic Adam optimizer that provides an automated, dynamic adjustment to the adaptive learning rate based on their detailed study into the effects of variance and momentum during training.”
- How to out a spy satellite with a Jupyter Notebook
- A curated list of applied machine learning and data science notebooks and libraries across different industries
- A Selective Overview of Deep Learning
“From the statistical and scientific perspective, it is natural to ask: What is deep learning? What are the new characteristics of deep learning, compared with classical methods? We highlight new characteristics of deep learning (including depth and over-parametrization) and explain their practical and theoretical benefits. We also sample recent results on theories of deep learning, many of which are only suggestive. While a complete understanding of deep learning remains elusive, we hope that our perspectives and discussions serve as a stimulus for new statistical research.”
- Unsupervised learning of landmarks by Descriptor Vector Exchange
“We develop a new perspective on the equivariance approach by noting that dense landmark detectors can be interpreted as local image descriptors equipped with invariance to intra-category variations.”
- Turbo, An Improved Rainbow Colormap for Visualization
“We are happy to introduce Turbo, a new colormap that has the desirable properties of Jet while also addressing some of its shortcomings, such as false detail, banding and color blindness ambiguity.”
- Those Hurricane Maps Don’t Mean What You Think They Mean
“We use hurricane forecasts to warn people. Why do we misinterpret them so often?”
- After 5,000 games, Microsoft’s Suphx AI can defeat top Mahjong players
Mahjong is what’s known as an imperfect information game, meaning a number of factors remain unknown to players throughout matches. For instance, they must account for opponents’ unseen tiles and decide whether to fold, leading to bluffs.
- Building an ML−enabled fullstack application with Vue, Flask, Mongo, and Algorithmia
- Simple Python Package to Extract Deep Learning Features
“Ever wanted to do a hacky computer vision project? But you don’t want to invest time on learning/using complicated deep learning libraries like PyTorch or TensorFlow? Enter image_features.”
“PySceneDetect is a command-line application and a Python library for detecting scene changes in videos, and automatically splitting the video into separate clips.”