Tree based learning algorithms are quite common in data science competitions. These algorithms empower predictive models with high accuracy, stability and ease of interpretation. Unlike linear models, they map non-linear relationships quite well. Common examples of tree based models are: decision trees, random forest, and boosted trees. In this post, we will look at the mathematical details (along with various python examples) of decision trees, its advantages and drawbacks. We will find that they are simple and very useful for interpretation. However, they typically are not competitive with the best supervised learning approaches.

There is now tremendous momentum behind initiatives to teach computer programming to a broad audience, yet many of these efforts (for example, Code.org,

By 2020, 25% of the American workforce will be over the age of 55 and approaching retirement, a phenomenon becoming known as the Silver Tsunami. While this could create a shortage of skilled workers in a number of fields including electric utilities, telecommunications, and manufacturing, augmented reality (AR) is poised not only to address issues faced by our aging workforce, but to fundamentality increase productivity by changing how all employees are trained in the future. In 2016, U.S. companies across industries spent nearly $1,000 in training per employee, largely delivered in traditional formats like classroom-based seminars and classes, and even online training modules that mimic that experience. This kind of learning has suited people's needs for centuries, particularly when learning was thought of as memorization with many cultures celebrating those who could recite long texts with exceptional rote skills. But as the breadth of human knowledge expanded, learning paradigms have changed with the works of John Dewey and others who recognized that understanding why information is important and how it relates to our world is true learning--and should be the goal.

But researchers in AI, and related fields such as learning analytics, are also thinking about how AI can provide more effective feedback to students and teachers. This is the use of technology – including AI – to provide people with information that helps them make better decisions and learn more effectively. So, for instance, rather than focusing on automating the grading of student essays, some researchers are focusing on how they can provide intelligent feedback to students that helps them better assess their own writing. Intelligence amplification helps counteract these concerns by keeping people in the loop.

In the first of the three leading-edge sessions on artificial intelligence at this year's ILTACon, folks will be hearing from Martin Tully, Co-Chair, of Akerman LLP's Data Law Practice, and Samuel Whitman, Mayer Brown's Knowledge Management Leader (see below for speaker bios). I will be moderating this panel along with the other two in the AI series at ILTACon 2017. At the end of the panel discussion those in attendance should be well on their way towards determining a communication strategy to educate their teams about AI technologies, have an understanding about real use-cases of AI in legal today, and lastly, have the basics down when it comes to understanding terms like natural language processing and machine learning. Have suggestions for questions I should ask Martin and Samuel? Send them my way via Twitter!

In recent years it has become popular to study machine learning problems in a setting of ordinal distance information rather than numerical distance measurements. By ordinal distance information we refer to binary answers to distance comparisons such as $d(A,B)

There have been a lot of digital "next big things" in education over the years--everything from the Apple IIe to online learning. The latest is artificial intelligence education tech (AI Ed), and only time will tell what impact it will ultimately have. But for something as important as education, now is the time to start talking about the benefits and challenges created by AI-powered personalized learning systems as they make their way into classrooms. Entefy covered this topic in previous articles: Old school no more: AI disrupts the classroom, which focused on teachers; and Artificial intelligence may transform education, but are parents ready?, which focused on parents. The clear near-term opportunity for AI Ed is to support teachers by taking over time-consuming, lower-value tasks like grading and recordkeeping. But there are already sophisticated AI teaching systems under development, systems that raise long-term questions about what place AI should have in schools.

Ophir Tanz is the CEO of GumGum, an artificial intelligence company with particular expertise in computer vision. GumGum applies its capabilities to a variety of industries, from advertising to professional sports across the globe. Ophir holds a B.S. and a M.S. from Carnegie Mellon University and currently lives in Los Angeles. Cambron Carter leads the image technology team at GumGum, where he designs computer vision and machine learning solutions for a wide variety of applications. Cambron holds B.S. degrees in physics and electrical engineering and an M.Eng. in electrical engineering from the University of Louisville.

Please check Piazza regularly for announcements and discussion. I will attempt to post slides before lecture. Readings in CIML are required. Other readings are optional unless otherwise specified. Domingos, Pedro Domingos' video lectures on Coursera There are many excellent machine learning textbooks, but none of them is quite perfect for this class.

The next time you come across a picture of a ravishing dish on Instagram or WeChat that whets your appetite but you can't exactly make out what it is made of, don't wrack your brain trying to guess the recipe. An AI system unwrapped earlier this week helps those with culinary curiosity find the right ingredients of an unknown dish and offers step-by-step instructions how to make it just by analyzing a photo they upload online. Researchers from the Polytechnic University of Catalonia, Massachusetts Institute of Technology (MIT) and Qatar Computing Research Institute have developed a deep-learning algorithm that can whip out a recipe just by "looking" at a photo of the dish. They fed the neural network one million recipes, along with one million photos of their final outcome, from popular websites like Allrecipes.com and Food.com to create a huge database they dubbed, Recipe1M, accessible through a web portal they called Pic2Recipe. With a single click of a button, the website allows users to upload a photo of the mystery dish and then the system, using machine learning, goes through the massive mounds of data to analyze it. It then predicts a list of possible ingredients along with their relevant recipes, then ranks them based on how certain the AI is they match the image.