The development of accurate and transferable machine learning (ML) potentials for predicting molecular energetics is a challenging task. The process of data generation to train such ML potentials is a task neither well understood nor researched in detail. In this work, we present a fully automated approach for the generation of datasets with the intent of training universal ML potentials. It is based on the concept of active learning (AL) via Query by Committee (QBC), which uses the disagreement between an ensemble of ML potentials to infer the reliability of the ensemble's prediction. QBC allows our AL algorithm to automatically sample regions of chemical space where the machine learned potential fails to accurately predict the potential energy. AL improves the overall fitness of ANAKIN-ME (ANI) deep learning potentials in rigorous test cases by mitigating human biases in deciding what new training data to use. AL also reduces the training set size to a fraction of the data required when using naive random sampling techniques. To provide validation of our AL approach we develop the COMP6 benchmark (publicly available on GitHub), which contains a diverse set of organic molecules. We show the use of our proposed AL technique develops a universal ANI potential (ANI-1x), which provides very accurate energy and force predictions on the entire COMP6 benchmark. This universal potential achieves a level of accuracy on par with the best ML potentials for single molecule or materials while remaining applicable to the general class of organic molecules comprised of the elements CHNO.

About this course: Leaders in business and non-profit organizations increasingly work across national borders and in multi-cultural environments. You may work regularly with customers or suppliers abroad, or be part of a globally dispersed cross-functional team, or an expatriate manager on an international assignment. You may be a member of a global online community, or a development aid worker collaborating with an international network of partner organizations. In all of these contexts, your effectiveness as a leader depends on how well you understand and are able to manage individual and collective behaviors in an intercultural context. In this course – together with a team of Bocconi expert faculty and Bocconi alumni – we'll explore the theory and practice of international and intercultural leadership and organizational behavior.

I have an upcoming interview that involves applying Deep Learning to Computer Vision problems. Though I have experience with deep learning I'm currently weak on the pure Computer Vision side of things. What are the topics that I should revise? What questions might be asked?

Bradbury taps into a human concern that surrounds the idea of Artificial Intelligence – if technology continues to develop at such a rate, humans will begin to become obsolete in our own homes and workplaces. Every week there's a new article telling us that robots can do our jobs better than we can, after all. But step back from the hysteria--the robots are not really coming to get us--and AI is already very much a part of our lives. Computers have been mimicking cognitive functions for many years: Deep Blue beat Kasparov in a chess match in 1996. We've all been helped (or hindered) by a chatbot online.

Neurons store and transmit information in the brain.Credit: CNRI/SPL Superconducting computing chips modelled after neurons can process information faster and more efficiently than the human brain. That achievement, described in Science Advances on 26 January1, is a key benchmark in the development of advanced computing devices designed to mimic biological systems. And it could open the door to more natural machine-learning software, although many hurdles remain before it could be used commercially. Artificial intelligence software has increasingly begun to imitate the brain. Algorithms such as Google's automatic image-classification and language-learning programs use networks of artificial neurons to perform complex tasks. But because conventional computer hardware was not designed to run brain-like algorithms, these machine-learning tasks require orders of magnitude more computing power than the human brain does.

Any gradient descent optimization requires to choose a learning rate. With deeper and deeper models, tuning that learning rate can easily become tedious and does not necessarily lead to an ideal convergence. We propose a variation of the gradient descent algorithm in the which the learning rate is not fixed. Instead, we learn the learning rate itself, either by another gradient descent (first-order method), or by Newton's method (second-order). This way, gradient descent for any machine learning algorithm can be optimized.

Last Thanksgiving, I watched my father-in-law evaluate over one hundred exams for the high school class he teaches on the U.S. government. They were mostly short answer questions: matching different provisions of the U.S. Constitution, and explaining the contents of the Bill of Rights. The grading was tedious and time consuming, and took him hour after hour during what should have been a holiday. I started to wonder whether there could be a faster way. Automatic computer grading could do exactly that, learning from previous answers and getting better as it goes -- and it is already being used in some universities and for large online courses (MOOCs).

Executives increasingly see investing in retraining and "upskilling" existing workers as an urgent business priority that companies, not governments, must lead on. The world of work faces an epochal transition. By 2030, according to the a recent McKinsey Global Institute report, Jobs lost, jobs gained: Workforce transitions in a time of automation, as many as 375 million workers--or roughly 14 percent of the global workforce--may need to switch occupational categories as digitization, automation, and advances in artificial intelligence disrupt the world of work. The kinds of skills companies require will shift, with profound implications for the career paths individuals will need to pursue. How big is that challenge?

Candidates hoping to land their dream job are increasingly being asked to play video games, with companies like Siemens, E.ON and Walmart filtering out hundreds of applicants before the interview stage based partly on how they perform. Played on either smartphones or computers, the games' designers say they can help improve workplace diversity, but there are questions over how informative the results really are. To the casual observer, many of the games might seem almost nonsensical. One series of tests by UK-based software house Arctic Shores includes a trial where the player must tap a button frantically to inflate balloons for a party without bursting them. In another, the candidate taps a logo matching the one displayed on screen, at an ever more blistering pace.

Booz Allen Hamilton has been at the forefront of strategy and technology for more than 100 years. Today, the firm provides management and technology consulting and engineering services to leading Fortune 500 corporations, governments, and not-for-profits across the globe. Booz Allen partners with public and private sector clients to solve their most difficult challenges through a combination of consulting, analytics, mission operations, technology, systems delivery, cybersecurity, engineering and innovation expertise. So you want to be a Data Scientist? Booz Allen Hamilton is no longer just a consulting firm.