During the christmas vacation of 2015, I got a motivational boost again and decided try out Kaggle. So I spent quite some time experimenting with various algorithms for their Homesite Quote Conversion, Otto Group Product Classification and Bike Sharing Demand contests. The main takeaway from this was the experience of iteratively improving the results by experimenting with the algorithms and the data. I learned to trust my logic when doing machine learning. If tweaking a parameter or engineering a new feature seems like a good idea logically, it's quite likely that it actually will help.

Is your organization using, or planning to adopt, machine learning? If so, please share your experiences and insights in this survey. And even if you have no plans to use machine learning, please take the survey anyway--we'd love to know why.

Researchers are developing an app which could help to prevent suicides by flagging those most at risk. Using a computer algorithm, it records conversations, analysing what people say and how they speak. By picking up on a range of subtle verbal and non-verbal cues, it can correctly classify if someone is suicidal with 93 per cent accuracy. At the heart of the app is a machine learning algorithm which classifies the person based on their responses. In an earlier study, researchers enrolled a mix of 379 patients, who were suicidal, diagnosed as mentally ill, or neither.

Whether it is a phobia of spiders or a traumatic event in the past, the effects of fear can reverberate through a person's life. But with the right training we can rid ourselves of them, according to a study. Researchers have found a method for tricking the brain into letting go of specific fears, which they claim could lead to new treatments for phobias and post-traumatic stress disorder (PTSD). By combining functional brain imaging with artificial intelligence, researchers have been able to zero in on memories related to fear and retrain them. For people suffering from phobias, a common course of treatment is aversion therapy, bringing someone into contact with their fear. In the case of spiders, a person could be gradually introduced to the arachnids through photos, images and eventually the real thing, learning that their fear is far greater than any actual risk a spider poses.

Two captive harbour porpoises called Freja and Sif have helped to reveal that porpoises --and probably all cetaceans -- consciously adjust their heart rate to suit the length of a planned dive. By doing this, the animals optimise the rate at which they consume oxygen beforehand to match the intended depth and length of their dive. "Until now, we knew that the heart rates of porpoises and cetaceans in general correlate with different dive factors, such as dive duration, depth and exercise," says Siri Elmegaard of Aarhus University in Denmark, who led the research. "Now we can conclude that harbour porpoises have cognitive control of their heart rate." The discovery might also provide another explanation for how exposure to loud noise from shipping, sonar or subsea exploration harms cetaceans and possibly triggers strandings.

A few months ago I demonstrated how to install the Keras deep learning library with a Theano backend. In today's blog post I provide detailed, step-by-step instructions to install Keras using a TensorFlow backend, originally developed by the researchers and engineers on the Google Brain Team. I'll also (optionally) demonstrate how you can integrate OpenCV into this setup for a full-fledged computer vision deep learning development environment. The first part of this blog post provides a short discussion of Keras backends and why we should (or should not) care which one we are using. From there I provide detailed instructions that you can use to install Keras with a TensorFlow backend for machine learning on your own system.

Google announced today that it will extend funding for AI research at the Montreal Institute for Learning Algorithms (MILA). The company will invest a total of C$4.5 million, or about US$3.4 million, to fund seven faculty members across the institute, as well as MILA faculty at the University of Montreal and McGill University. That funding will also continue to support Yoshua Bengio, one of the few deep-learning experts currently working. There's no doubt that Google is hoping to expand its AI and deep-learning expertise by investing in folks like Benigo. Google also announced it will be opening up a deep-learning and AI research group at its offices in Montreal.

The saying goes: 'Never judge a book by its cover,' but that's exactly what new AI technology has been designed to do. A controversial paper has been released, which investigates whether a computer can detect if a human could be a criminal, by analysing their facial features. The results suggest that it is bad news for people with smaller mouths, curvier upper lips and closer-set eyes, as apparently these features suggest you could be a crook. The paper investigates whether a computer can detect if a human could be a criminal, by analysing their facial features. The researchers singled out three features that they suggest can tell whether someone will be a criminal or not – lip curvature, eye inner corner distance, and the angle from the tip of the nose to the corners of the mouth.

CONVERSATIONS on the future have seen a common theme emerge - that it is disrupted and predominantly digital. Technological advancements in artificial intelligence, robotics, sharing platforms and the Internet of Things are fundamentally altering business models and industries. These changes are often not only alien to businesses; they are taking place at unprecedented speed. Many organisations, in particular the larger established ones that are encumbered by complex structures and bureaucratic processes, may find adapting to the pace of change a real struggle. Entrepreneurial businesses are often said to have a huge competitive advantage by virtue of their smaller size and simpler business model, which affords them the needed agility.

Are art and science of distinctly different cultures? The former often seems fixated on human experience, the latter on physical processes. In his most recent book, Reductionism in Art and Brain Science: Bridging the Two Cultures, published this year, the Nobel Prize-winning neuroscientist Eric Kandel argues that such a separation no longer exists. The best-known abstractionists, like Mark Rothko, Jackson Pollock, Dan Flavin, and Willem de Kooning, Kandel writes, effectively created "new rules for visual processing." Abstract art, says Kandel, is therefore the key to understanding both how art and science inform one another, and together, they might open up entirely new ways of seeing and imagining.