Whether you realize it or not, artificial intelligence (AI) is already shaping our world. Whenever you use Siri or Alexa, you are already communicating with digital assistance – a form of AI that is pegged as an "intelligent digital assistant." These AI assistants are designed to make your life easier. Now, it is clear they are headed to the classroom as well. According to Artificial Intelligence Market in the US Education Sector 2017-2021, experts expect AI in education to grow by "47.50% during the period 2017-2021."

Siri and Alexa are clearly far from perfect, but there is hope that steady progress in machine learning will turn them into articulate helpers before long. A new test, however, may help show that a fundamentally different approach is required for AI systems to actually master language. Developed by researchers at the Allen Institute for AI (AI2), a nonprofit based in Seattle, the AI2 Reasoning Challenge (ARC) will pose elementary-school-level multiple-choice science questions. Each question will require some understanding of how the world works. The project is described in a related research paper (pdf).

We present a neural model for representing snippets of code as continuous distributed vectors. The main idea is to represent code as a collection of paths in its abstract syntax tree, and aggregate these paths, in a smart and scalable way, into a single fixed-length \emph{code vector}, which can be used to predict semantic properties of the snippet. We demonstrate the effectiveness of our approach by using it to predict a method's name from the vector representation of its body. We evaluate our approach by training a model on a dataset of $14$M methods. We show that code vectors trained on this dataset can predict method names from files that were completely unobserved during training. Furthermore, we show that our model learns useful method name vectors that capture semantic similarities, combinations, and analogies. Comparing previous techniques over the same data set, our approach obtains a relative improvement of over $75\%$, being the first to successfully predict method names based on a large, cross-project, corpus.

Developing state-of-the-art approaches for specific tasks is a major driving force in our research community. Depending on the prestige of the task, publishing it can come along with a lot of visibility. The question arises how reliable are our evaluation methodologies to compare approaches? One common methodology to identify the state-of-the-art is to partition data into a train, a development and a test set. Researchers can train and tune their approach on some part of the dataset and then select the model that worked best on the development set for a final evaluation on unseen test data. Test scores from different approaches are compared, and performance differences are tested for statistical significance. In this publication, we show that there is a high risk that a statistical significance in this type of evaluation is not due to a superior learning approach. Instead, there is a high risk that the difference is due to chance. For example for the CoNLL 2003 NER dataset we observed in up to 26% of the cases type I errors (false positives) with a threshold of p < 0.05, i.e., falsely concluding a statistically significant difference between two identical approaches. We prove that this evaluation setup is unsuitable to compare learning approaches. We formalize alternative evaluation setups based on score distributions.

The k-Nearest-Neighbors (kNN) method of classification is one of the simplest methods in machine learning, and is a great way to introduce yourself to machine learning and classification in general. At its most basic level, it is essentially classification by finding the most similar data points in the training data, and making an educated guess based on their classifications. Although very simple to understand and implement, this method has seen wide application in many domains, such as in recommendation systems, semantic searching, and anomaly detection. As we would need to in any machine learning problem, we must first find a way to represent data points as feature vectors. A feature vector is our mathematical representation of data, and since the desired characteristics of our data may not be inherently numerical, preprocessing and feature-engineering may be required in order to create these vectors.

The shooting at the school in Florida was devastating, and it appears clear that Russia has been manipulating public opinion in the U.S. to stoke the flames of a divisive argument on guns. What is being missed is a brewing problem that potentially could have an even more devastating impact. Competing for our eyeballs is the news that the U.S. president kissed a woman without her permission. That story has served as a distraction from far more horrendous attacks against women in the tech industry and in government. Intel showcased virtual reality at the Olympics, but almost no one cared.

The more I've had the opportunity to explore this concept of the martech mindset, the more I've realized how important it is that we shift our thinking to view artificial intelligence (AI) as an evolution rather than as a revolution. Artificial intelligence, once an abstract, futuristic concept, is now a reality, so this needs to occur sooner rather than later. It's no longer a hypothetical; AI is changing the game for brands in tangible, practical applications. Adoption is not the issue, as investments in AR and VR are expected to grow from $11.4 billion in 2017 to $215 billion in 2021, according to IDC. Entire industries are finding innovative ways to solve persistent business challenges and create efficiencies through AI -- take the proliferation of chatbots in marketing, for example.

Over the last decade, Mexican drug cartels have been fighting each other--and corrupt police and military units--for control of the lucrative drug trade, plunging the country into chaos. Outsiders might think of Mexico as sunny and tequila-soaked, but beyond the beach resorts of Cancun and Mazatlan there hides a grimmer tale: levels of murder, rape, and kidnapping are hitting levels rarely seen outside hotspots in Africa, Asia, and South America. So grim the tale, when 43 college students went missing in Mexico's southern state of Guerrero in 2014, investigators found 129 other bodies in 60 fosas clandestinas (mass graves) before stumbling on badly burned remains in a mass grave they think might--possibly, maybe--contain what's left of the missing students. Mexico's attorney general says the local mayor conspired with the town's police force to abduct the students and turn them over to a local gang, who murdered them and burned the bodies, and dumped the charred corpses into a river. The situation is so bad that, after six decades of gains, the average life expectancy in Mexico has decreased, according to recent research.

Learning linear predictors with the logistic loss---both in stochastic and online settings---is a fundamental task in learning and statistics, with direct connections to classification and boosting. Existing "fast rates" for this setting exhibit exponential dependence on the predictor norm, and Hazan et al. (2014) showed that this is unfortunately unimprovable. Starting with the simple observation that the logistic loss is 1-mixable, we design a new efficient improper learning algorithm for online logistic regression that circumvents the aforementioned lower bound with a regret bound exhibiting a doubly-exponential improvement in dependence on the predictor norm. This provides a positive resolution to a variant of the COLT 2012 open problem of McMahan and Streeter (2012) when improper learning is allowed. This improvement is obtained both in the online setting and, with some extra work, in the batch statistical setting with high probability. We also show that the improved dependency on predictor norm is also near-optimal. Leveraging this improved dependency on the predictor norm yields the following applications: (a) we give algorithms for online bandit multiclass learning with the logistic loss with an $\tilde{O}(\sqrt{n})$ relative mistake bound across essentially all parameter ranges, thus providing a solution to the COLT 2009 open problem of Abernethy and Rakhlin (2009), and (b) we give an adaptive algorithm for online multiclass boosting with optimal sample complexity, thus partially resolving an open problem of Beygelzimer et al. (2015) and Jung et al. (2017). Finally, we give information-theoretic bounds on the optimal rates for improper logistic regression with general function classes, thereby characterizing the extent to which our improvement for linear classes extends to other parameteric and even nonparametric settings.

Recent advances in policy gradient methods and deep learning have demonstrated their applicability for complex reinforcement learning problems. However, the variance of the performance gradient estimates obtained from the simulation is often excessive, leading to poor sample efficiency. In this paper, we apply the stochastic variance reduced gradient descent (SVRG) to model-free policy gradient to significantly improve the sample-efficiency. The SVRG estimation is incorporated into a trust-region Newton conjugate gradient framework for the policy optimization. On several Mujoco tasks, our method achieves significantly better performance compared to the state-of-the-art model-free policy gradient methods in robotic continuous control such as trust region policy optimization (TRPO)