We consider regret minimization in repeated games with non-convex loss functions. Minimizing the standard notion of regret is computationally intractable. Thus, we define a natural notion of regret which permits efficient optimization and generalizes offline guarantees for convergence to an approximate local optimum. We give gradient-based methods that achieve optimal regret, which in turn guarantee convergence to equilibrium in this framework.

It branches out from Artificial Intelligence and is classified as a Machine Learning type. Leveraging reinforcement learning, software agents and machines are made to ascertain the ideal behavior in a specific context with the aim of maximizing its performance. When the learning agent acts on trial and error, it is termed as exploration, and when it acts based on the knowledge gained from the environment, it is referred to as exploitation. The environment rewards the agent for correct actions, which is the reinforcement signal. Leveraging the rewards obtained, the agent improves its environment knowledge to select the next action.

So far I have talked about decision trees and ensembles. But I hope, I have made you understand the logic behind these concepts without getting too much into the mathematical details. In this post lets get into action, I will be implementing the concepts that we learned in these two blog posts. The only concept that I haven't discussed about is SVM. I suggest you to watch Professor Andrew Ng's week 7 videos on Coursera.

Last Friday, we mentioned how Google's artificial intelligence software DeepMind has the ability to teach itself many things. It can teach itself how to walk, jump and run. Or defeat the world's best player of the Chinese strategy game, Go. The science of teaching computers how to do things is called Deep Learning. Offered through Udacity, the course is taught by Vincent Vanhoucke, the technical lead in Google's Brain team.

Don't doubt it: Machine learning is hot--and getting hotter. For the past two years, public interest in building complex algorithms that automatically "learn" and improve from their own operations, or experience (rather than explicit programming) has been growing. Call it "artificial intelligence," or (better) "machine learning." Such work has, in fact, been going on for decades. More recently, Shivon Zilis, an investor with Bloomberg Beta, has been building a landscape map of where machine learning is being applied across other industries.

Machine learning is exploding, with smart algorithms being used everywhere from email to smartphone apps to marketing campaigns. Translation: if you're looking for an in-demand career, setting yourself up with the skills to work with smart machines/artificial intelligence is a good move. With input from Florian Douetteau, CEO of Dataiku, here are some things you can start doing today to position yourself for a future career in machine learning. This may sound obvious, says Douetteau, but it's important. "Having experience and understanding of what machine learning is, understanding the basic maths behind it, understanding the alternative technology, and having experience -- hands-on experience -- with the technology is key."

FOR a ten-year-old, Amartya is a thoughtful chap. One Monday morning at the Khan Lab School (KLS) in Mountain View, California, he explains that his maths is "pretty strong" but he needs to work on his writing. Not to worry, though; Amartya has a plan. He will practise grammar online, book a slot with an English teacher and consult his mentor. Later he will e-mail your correspondent to ask for help, too. This is the sort of pluck KLS produces. Its pupils do not have homework or report cards or spend all day in classrooms.

When Ross Compton had a pacemaker installed, he had a constitutional right to remain silent. One would expect his body to have the same. But when the 59-year-old's Middletown, Ohio, home erupted in flames last September, the electronic data stored in his cardiac device eventually led to his arrest and subsequent indictment on charges of arson and insurance fraud. And despite his attorney's arguments to the contrary, earlier this month Butler county judge Charles Pater held that the functioning of Compton's own body -- heartbeat included -- could be used against him at the upcoming trial. Deanna Paul (@thedeannapaul) is a former New York City prosecutor and adjunct professor of trial advocacy at Fordham University school of law.

We study the active learning problem of top-$k$ ranking from multi-wise comparisons under the popular multinomial logit model. Our goal is to identify the top-$k$ items with high probability by adaptively querying sets for comparisons and observing the noisy output of the most preferred item from each comparison. To achieve this goal, we design a new active ranking algorithm without using any information about the underlying items' preference scores. We also establish a matching lower bound on the sample complexity even when the set of preference scores is given to the algorithm. These two results together show that the proposed algorithm is nearly instance optimal (similar to instance optimal [FLN03], but up to polylog factors). Our work extends the existing literature on rank aggregation in three directions. First, instead of studying a static problem with fixed data, we investigate the top-$k$ ranking problem in an active learning setting. Second, we show our algorithm is nearly instance optimal, which is a much stronger theoretical guarantee. Finally, we extend the pairwise comparison to the multi-wise comparison, which has not been fully explored in ranking literature.

Are our machines turning into gods? Jie, the world's best player of the world's oldest board game, Go, had just met his match… in the form of a program called AlphaGo. In the space of a year, the program had become "almost like the god of Go," said Jie after losing to AlphaGo. Jie had been playing the game, viewed as too hard for machines to excel at, since he was 10. AlphaGo was only made by Google's parent, Alphabet, in 2014.