In the cosmic scale of things, black holes are a dime a dozen. Despite this, and despite what Christopher Nolan's sci-fi blockbuster "Interstellar" would have you believe, we humans have never actually seen one with our eyes. A team of researchers from the Massachusetts Institute of Technology's artificial intelligence laboratory and the Harvard University revealed Monday that they had developed an algorithm that may allow us to actually "see" black holes. "We would never be able to see into the center of our galaxy in visible wavelengths because there's too much stuff in between," Katie Bouman, an MIT graduate student in electrical engineering and computer science who led the development of the new algorithm, said in a statement. "A black hole is very, very far away and very compact. To image something this small means that we would need a telescope with a 10,000-kilometer diameter, which is not practical, because the diameter of the Earth is not even 13,000 kilometers."

The ALMA telescope array is one of many groups involved with the Event Horizon Telescope project which hopes to image black holes. Humans have never actually seen a black hole with our eyes. A new computer algorithm, created by MIT graduate student Katie Bouman and her team, will hopefully change that. Since black holes are surprisingly small, we need a gigantic telescope to look at them. The supermassive black hole at the center of our Milky Way galaxy is only about 17 times the diameter of the sun, according to an article published in BBC News, but it's 25,000 light years away... that's far.

The most common question asked by prospective data scientists is – "What is the best programming language for Machine Learning?" The answer to this question always results in a debate whether to choose R, Python or MATLAB for Machine Learning. Nobody can, in reality, answer the question as to whether Python or R is best language for Machine Learning. However, the programming language one should choose for machine learning directly depends on the requirements of a given data problem, the likes and preferences of the data scientist and the context of machine learning activities they want to perform. According to a survey on Kaggler's Favourite Tools, the open source R programming language turned out to be the favourite among 543 Kagglers of the 1714 Kaggler's listing their data science tools.

In the study of human learning, there is broad evidence that our ability to retain information improves with repeated exposure and decays with delay since last exposure. This plays a crucial role in the design of educational software, leading to a trade-off between teaching new material and reviewing what has already been taught. A common way to balance this trade-off is spaced repetition, which uses periodic review of content to improve long-term retention. Though spaced repetition is widely used in practice, e.g., in electronic flashcard software, there is little formal understanding of the design of these systems. Our paper addresses this gap in three ways. First, we mine log data from spaced repetition software to establish the functional dependence of retention on reinforcement and delay. Second, we use this memory model to develop a stochastic model for spaced repetition systems. We propose a queueing network model of the Leitner system for reviewing flashcards, along with a heuristic approximation that admits a tractable optimization problem for review scheduling. Finally, we empirically evaluate our queueing model through a Mechanical Turk experiment, verifying a key qualitative prediction of our model: the existence of a sharp phase transition in learning outcomes upon increasing the rate of new item introductions.

We study the problem of structured prediction under test-time budget constraints. We propose a novel approach applicable to a wide range of structured prediction problems in computer vision and natural language processing. Our approach seeks to adaptively generate computationally costly features during test-time in order to reduce the computational cost of prediction while maintaining prediction performance. We show that training the adaptive feature generation system can be reduced to a series of structured learning problems, resulting in efficient training using existing structured learning algorithms. This framework provides theoretical justification for several existing heuristic approaches found in literature. We evaluate our proposed adaptive system on two structured prediction tasks, optical character recognition (OCR) and dependency parsing and show strong performance in reduction of the feature costs without degrading accuracy.

Iterative procedures for parameter estimation based on stochastic gradient descent allow the estimation to scale to massive data sets. However, in both theory and practice, they suffer from numerical instability. Moreover, they are statistically inefficient as estimators of the true parameter value. To address these two issues, we propose a new iterative procedure termed averaged implicit SGD (AI-SGD). For statistical efficiency, AI-SGD employs averaging of the iterates, which achieves the optimal Cram\'{e}r-Rao bound under strong convexity, i.e., it is an optimal unbiased estimator of the true parameter value. For numerical stability, AI-SGD employs an implicit update at each iteration, which is related to proximal operators in optimization. In practice, AI-SGD achieves competitive performance with other state-of-the-art procedures. Furthermore, it is more stable than averaging procedures that do not employ proximal updates, and is simple to implement as it requires fewer tunable hyperparameters than procedures that do employ proximal updates.

We introduce the Adaptive Skills, Adaptive Partitions (ASAP) framework that (1) learns skills (i.e., temporally extended actions or options) as well as (2) where to apply them. We believe that both (1) and (2) are necessary for a truly general skill learning framework, which is a key building block needed to scale up to lifelong learning agents. The ASAP framework can also solve related new tasks simply by adapting where it applies its existing learned skills. We prove that ASAP converges to a local optimum under natural conditions. Finally, our experimental results, which include a RoboCup domain, demonstrate the ability of ASAP to learn where to reuse skills as well as solve multiple tasks with considerably less experience than solving each task from scratch.

Online video games are to blame for a decline in young men entering higher education. This is the neat claim made in a recent op-ed published by the Times under the subheading "The gender imbalance in higher education may not be as complicated as it looks". Emboldened by a recent report from the Higher Education Policy Institute, which found that boys were 10 times as likely to play collaborative online games than girls, the author repeats its assertion that "the gender gap in video gaming translates into a performance advantage for girls". Not correlation, then, but grim causation: play video games, drop grades. The gender gap in higher education in the UK is growing.

Algorithm design is a laborious process and often requires many iterations of ideation and validation. In this paper, we explore automating algorithm design and present a method to learn an optimization algorithm, which we believe to be the first method that can automatically discover a better algorithm. We approach this problem from a reinforcement learning perspective and represent any particular optimization algorithm as a policy. We learn an optimization algorithm using guided policy search and demonstrate that the resulting algorithm outperforms existing hand-engineered algorithms in terms of convergence speed and/or the final objective value.

We present a generic compact computational framework relying on structured random matrices that can be applied to speed up several machine learning algorithms with almost no loss of accuracy. The applications include new fast LSH-based algorithms, efficient kernel computations via random feature maps, convex optimization algorithms, quantization techniques and many more. Certain models of the presented paradigm are even more compressible since they apply only bit matrices. This makes them suitable for deploying on mobile devices. All our findings come with strong theoretical guarantees. In particular, as a byproduct of the presented techniques and by using relatively new Berry-Esseen-type CLT for random vectors, we give the first theoretical guarantees for one of the most efficient existing LSH algorithms based on the $\textbf{HD}_{3}\textbf{HD}_{2}\textbf{HD}_{1}$ structured matrix ("Practical and Optimal LSH for Angular Distance"). These guarantees as well as theoretical results for other aforementioned applications follow from the same general theoretical principle that we present in the paper. Our structured family contains as special cases all previously considered structured schemes, including the recently introduced $P$-model. Experimental evaluation confirms the accuracy and efficiency of TripleSpin matrices.