Which authors of this paper are endorsers? Disable MathJax (What is MathJax?)

Data science today is a lot like the Wild West: there's endless opportunity and excitement, but also a lot of chaos and confusion. If you're new to data science and applied machine learning, evaluating a machine-learning model can seem pretty overwhelming. With this O'Reilly report, machine-learning expert Alice Zheng takes you through the model evaluation basics. In this overview, Zheng first introduces the machine-learning workflow, and then dives into evaluation metrics and model selection. The latter half of the report focuses on hyperparameter tuning and A/B testing, which may benefit more seasoned machine-learning practitioners.

A First Course in Machine Learning by Simon Rogers and Mark Girolami is the best introductory book for ML currently available. It combines rigor and precision with accessibility, starts from a detailed explanation of the basic foundations of Bayesian analysis in the simplest of settings, and goes all the way to the frontiers of the subject such as infinite mixture models, GPs, and MCMC.

Artificial intelligence is slowly seeping in through current techs. Recently, Microsoft has published the source code for its Project Malmo. Project Malmo is a tool that will allow anyone to make artificial intelligence experiments when it comes to Minecraft games based on the programming skills provided. According to Neuro Gadget, the project was previously announced back in March but it was named as AIX. Now, with a new name, the company released it for the public in order for people to test it out.

Cisco announced a partnership Monday with bot-building platforms Gupshup and API.ai that allows thousands of bots to quickly join Cisco Spark and Cisco Tropo platforms. It also turns up the intensity in competition between enterprise team communication chat apps like Skype and Slack. The announcement was made during Cisco Live, a four-day Cisco event taking place in Las Vegas this week. Gupshup built an SMS social network of more than 50 million users, mainly in India, before becoming an enterprise messaging service company in 2010. Today, it processes four billion messages a month.

To learn more about opportunities in applied AI, join us at the O'Reilly Artificial Intelligence Conference, September 26-27, 2016 in New York. One of the truisms of modern AI is that the next big step is to move from supervised to unsupervised learning. In the last few years, we've made tremendous progress in supervised learning: photo classification, speech recognition, even playing Go (which represents a partial, but only partial, transition to unsupervised learning). Unsupervised learning is still an unsolved problem. As Yann LeCun says, "We need to solve the unsupervised learning problem before we can even think of getting to true AI."

It always has, since the beginning of my forays into Computer Science. I did two startups in the nineties, at the heart of which were AI-driven innovations. The world's recent embrace of AI, thus, also fascinates me. Driverless cars are all the rage these days. Uber is chomping at the bit to replace the headache of having to pay anything at all to their drivers, having driven the labor cost down already to the minimum.

Neural networks are changing the Internet. Inspired by the networks of neurons inside the human brain, these deep mathematical models can learn discrete tasks by analyzing enormous amounts of data. They've learned to recognize faces in photos, identify spoken commands, and translate text from one language to another. They're helping to choose what you see when you query the Google search engine or visit your Facebook News Feed. All this is sharpening the behavior of online services.

Carmakers and tech companies are in a race to put autonomous vehicles on the road, and it's time for regulators to tap the brakes. This month the National Highway Traffic Safety Administration revealed that it is investigating two crashes involving Tesla vehicles allegedly operating on autopilot. Tesla's autopilot feature is a semi-autonomous system that uses cameras, radar and sensors to steer the car, change lanes, adjust speed and even find a parking space and parallel park. It's not supposed to turn a Tesla sedan into a self-driving car, but there's ample evidence on YouTube of people driving with their hands off the steering wheel, playing games and even climbing into the back seat while their car is hurtling down a freeway. You and your daughter are riding in a driverless car along Pacific Coast Highway.

Tech giant Google has added to its vast technology services stable by acquiring Paris-based startup Moodstocks, which has developed machine learning based image recognition technology for smartphones. The acquisition by Google will see it add the recognition technology of Moodstocks to its already large number of services it offers which use machine learning, such as Google Translate, Smart Reply in Inbox and the Goggle app. With many of its services already relying on machine learning technologies, Google's acquisition of Moodstocks will help with implementing visual recognition. The Moodstocks team of engineers and researchers, based in Paris, developed new algorithms for Visual pattern recognition and machine learning, as well as a technology for the recognition of images and objects via mobile devices. Head of the R&D Center of Google France, Vincent Simonet, wrote in a blog on July 7 to announce the deal, said that while great steps forward were taken by Google in terms of Visual recognition, there was still much to be done in this area, stating that the company expects that to be where Moodstocks comes into its own.