News

Germany adopts first ethics standards for autonomous driving systems

Federal transport minister, Alexander Dobrindt, presented a report to Germany’s cabinet seeking to establish guidelines for the future programming of ethical standards into automated driving software. The report, was prepared by an automated driving ethics commission comprised of scientists and legal experts and produced 20 guidelines to be used by the automotive industry when creating automated driving systems. Shortly after its introduction, Dobrindt announced that the cabinet ratified the guidelines, making Germany the first government in the world to put such measures in place.

Self-repairing organs could save your life in a heartbeat

WHAT becomes of the broken-hearted? In cardiac medicine, the answer is usually brutally straightforward: they die. Heart disease is the leading cause of death worldwide and there is often precious little we can do about it.

‘Double Trojan Horse’ Drug Tricks Bacteria into Committing Suicide

Bacteria come in two very broad categories based on the structure of their cell walls, the outer region that gives the cells shape and integrity. The cell walls of Gram-positive bacteria consist of a membrane surrounded by a thick layer of sugar and protein, while the cell walls of Gram-negative bacteria consist of a membrane surrounded by a second membrane. This fundamental anatomical difference has a profound medical implication: The types of antibiotics that can kill Gram-positive bacteria are likely ineffective against Gram-negative bacteria and vice versa.

Automatic Photography with Google Clips

How could we train an algorithm to recognize interesting moments? As with most machine learning problems, we started with a dataset. We created a dataset of thousands of videos in diverse scenarios where we imagined Clips being used.

450,000 Women Missed Breast Cancer Screenings Due to “Algorithm Failure”

Custom deep learning loss functions with Keras for R

I recently started reading “Deep Learning with R”, and I’ve been really impressed with the support that R has for digging into deep learning. One of the use cases presented in the book is predicting prices for homes in Boston, which is an interesting problem because homes can have such wide variations in values. This is a machine learning problem that is probably best suited for classical approaches, such as XGBoost, because the data set is structured rather than perceptual data.

Nasa will send helicopter to Mars to test otherworldly flight

Apple made more profit in three months than Amazon has generated in its lifetime

On Cumulative Advantage and How to Think About Luck

In the late 1970s the view in the publishing world was that an author should never produce more than one book a year. The thinking was that publishing more than one book a year would dilute the brand name of the author. However, this was a bit of a problem for Stephen King, who was writing books at a rate of two per year.

AI Can Generate ‘Doom’ Levels Now

Researchers recently successfully trained neural networks to generate level maps for Doom that, they report in a paper published to the arXiv preprint server in April, “proved to be interesting” to play. The work was carried out by researchers from the Polytechnic University of Milan and used Generative Adversarial Networks, a recent innovation in the field of deep learning.