As part of the IEEE RAS International Conference on Humanoid Robots in Birmingham, U.K., last month, the awards committee decided to organize a fun photo contest. Participants submitted 39 photos showing off their humanoids in all kinds of poses and places. I was happy to be one of the judges, along with Sabine Hauert from the University of Bristol and Robohub, and with Giorgio Metta, the conference's awards chair, overseeing our selection. All photos were posted on Facebook and Twitter, and users were invited to vote on them. Sabine and I then looked at the photos with the most votes and scored them for originality, creativity, photo structure, and tech or fun factor.

Personalized learning, which tailors educational content to the unique needs of individual students, has become a huge component of K–12 education. A growing number of college educators are embracing the trend, taking advantage of data analytics and artificial intelligence to deliver just-right, just-in-time learning to their students. Data-driven insights are becoming integral to business and financial decision-making by institutional leaders, and educators are quickly finding ways to leverage analytics to increase student retention. Applying data analytics to adaptive learning programs is proving to be another smart application. In adaptive learning, educators collect data on various aspects of student performance -- from engagement with course content to exam performance -- and tailor material to each student's knowledge level and ideal learning style.

Technology is constantly evolving and adapting to boost everyday efficiency and make our lives easier. Modern tools give us the power to connect from around the globe and bridge gaps as soon as they appear. One such advancement is the rise of Machine Learning and Artificial Intelligence. Predictions, algorithms, and analytics come together to create more personalized eLearning experiences. But how exactly will Machine Learning and Artificial intelligence (AI) transform the eLearning landscape in years to come?

Machine Learning is one of the most exciting fields in the world. Every week we discover something new, something amazing, something revolutionary. That's why we created This Week in Machine Learning! Each week we publish a curated list of Machine Learning stories as a resource to help you keep pace with all these exciting developments. New posts will be published here first, and previous posts are archived on the Udacity blog.

In this setting, we only have access to the stochastic function f(x; ξ), where the random variable ξ is sampled from an underlying distribution D. The task is to optimize the expected function f(x), which in general may be nonconvex. This framework covers a wide range of problems, including the offline setting where we minimize the empirical loss over a fixed amount of data, and the online setting where data arrives sequentially. One of the most prominent applications of stochastic optimization has been in large-scale statistics and machine learning problems, such as the optimization of deep neural networks. Classical analysis in nonconvex optimization only guarantees convergence to a first-order stationary point (i.e., a point x satisfying ‖ f(x)‖ 0), which can be a local minimum, a local maximum, or a saddle point. This paper goes further, proposing an algorithm that escapes saddle points and converges to a local minimum.

Given a vertex of interest in a network $G_1$, the vertex nomination problem seeks to find the corresponding vertex of interest (if it exists) in a second network $G_2$. Although the vertex nomination problem and related tasks have attracted much attention in the machine learning literature, with applications to social and biological networks, the framework has so far been confined to a comparatively small class of network models, and the concept of statistically consistent vertex nomination schemes has been only shallowly explored. In this paper, we extend the vertex nomination problem to a very general statistical model of graphs. Further, drawing inspiration from the long-established classification framework in the pattern recognition literature, we provide definitions for the key notions of Bayes optimality and consistency in our extended vertex nomination framework, including a derivation of the Bayes optimal vertex nomination scheme. In addition, we prove that no universally consistent vertex nomination schemes exist. Illustrative examples are provided throughout.

Collected from the English exams for middle and high school Chinese students in the age range between 12 to 18, RACE consists of near 28,000 passages and near 100,000 questions generated by human experts (English instructors), and covers a variety of topics which are carefully designed for evaluating the students' ability in understanding and reasoning. In particular, the proportion of questions that requires reasoning is much larger in RACE than that in other benchmark datasets for reading comprehension, and there is a significant gap between the performance of the state-of-the-art models (43%) and the ceiling human performance (95%). We hope this new dataset can serve as a valuable resource for research and evaluation in machine comprehension.

Evaluating a policy by deploying it in the real world can be risky and costly. Off-policy policy evaluation (OPE) algorithms use historical data collected from running a previous policy to evaluate a new policy, which provides a means for evaluating a policy without requiring it to ever be deployed. Importance sampling is a popular OPE method because it is robust to partial observability and works with continuous states and actions. However, the amount of historical data required by importance sampling can scale exponentially with the horizon of the problem: the number of sequential decisions that are made. We propose using policies over temporally extended actions, called options, and show that combining these policies with importance sampling can significantly improve performance for long-horizon problems. In addition, we can take advantage of special cases that arise due to options-based policies to further improve the performance of importance sampling. We further generalize these special cases to a general covariance testing rule that can be used to decide which weights to drop in an IS estimate, and derive a new IS algorithm called Incremental Importance Sampling that can provide significantly more accurate estimates for a broad class of domains.

Observers are trained to observe children through their peripheral vision so that a child is unaware that he or she is being observed. Observers look at every child, one at a time, in a specified order. As soon as a child is showing a clear behavior, whether on or off task, it is noted and the observer moves on to the next child on his list. More than a dozen observations are taken for each child during each classroom session. This gives equal weight to all the children in the class and avoids overemphasizing attention-grabbing behaviors or highly distractable children.

Sophia, who became a citizen of Saudi Arabia in October, has a face that can show expression and metal hands. Sophia's clear "skull" shows the inner working wires of the artificially intelligent brain, which functions through a Wi-Fi connection pumped with information and a cohesive vocabulary.