Few can deny that there will be huge social, political, and economic transformation as a result of AI, robotics, and digital technologies. But there are many uncertainties about how those changes will play out, especially in the critical area of economics. Does it mean the end of work? From a historical perspective, we have been at similar crossroads before, MIT Economics Professor, Daron Acemoglu, said at the MIT IDE Annual Conference on May 24. Speaking about Automation and the Future of Work, Acemoglu said that many 20th-Century economists -- including John Maynard Keynes and Wassily Leontief -- also worried about technological unemployment in the 1930s and 1950s.

As scientific datasets increase in both size and complexity, the ability to label, filter and search this deluge of information has become a laborious, time-consuming and sometimes impossible task, without the help of automated tools. With this in mind, a team of researchers from Lawrence Berkeley National Laboratory (Berkeley Lab) and UC Berkeley are developing innovative machine learning tools to pull contextual information from scientific datasets and automatically generate metadata tags for each file. Scientists can then search these files via a web-based search engine for scientific data, called Science Search, that the Berkeley team is building. As a proof-of-concept, the team is working with staff at the Department of Energy's (DOE) Molecular Foundry, located at Berkeley Lab, to demonstrate the concepts of Science Search on the images captured by the facility's instruments. A beta version of the platform has been made available to Foundry researchers.

The trash truck rumbles down the street, and its cameras pour video into the city's data lake. An AI-powered application mines that image data looking for graffiti--and advises whether to dispatch a fully equipped paint crew or a squad with just soap and brushes. Meanwhile, cameras on other city vehicles could feed the same data lake so another application detects piles of trash that should be collected. That information is used by an application to send the right clean-up squad. Citizens, too, can get into the act, by sending cell phone pictures of graffiti or litter to the city for AI-driven processing.

In a world dominated by data, Data Science is the ladder to building a promising career in unique and challenging job positions. Kickstarting your career in Data Science is now easier than ever thanks to the vast pool of online platforms offering Data Science courses. These courses are specially designed to walk you through the concepts and intricacies of Data Science. But, do you know the exact way to climb the ladder? Fret not, for we're here to show you how! So, let's begin, shall we?

IMAGE: This is a screenshot of the Science Search interface. In this case, the user did an image search of nanoparticles. As scientific datasets increase in both size and complexity, the ability to label, filter and search this deluge of information has become a laborious, time-consuming and sometimes impossible task, without the help of automated tools. With this in mind, a team of researchers from Lawrence Berkeley National Laboratory (Berkeley Lab) and UC Berkeley are developing innovative machine learning tools to pull contextual information from scientific datasets and automatically generate metadata tags for each file. Scientists can then search these files via a web-based search engine for scientific data, called Science Search, that the Berkeley team is building.

IBM developer Dale Lane, who helped create the educational tool Machine Learning for Kids, believes that while the "most critical aspect of AI education is helping teachers to improve their own skills and educate our children more effectively", this continues to be overlooked. He shares Professor Luckin's frustration over the lack of progress so far.

In this case, the user performed an image search for nanoparticles. As scientific datasets increase in both size and complexity, the ability to label, filter and search this deluge of information has become a laborious, time-consuming and sometimes impossible task, without the help of automated tools. With this in mind, a team of researchers from the Department of Energy's Lawrence Berkeley National Laboratory (Berkeley Lab) and UC Berkeley are developing innovative machine learning tools to pull contextual information from scientific datasets and automatically generate metadata tags for each file. Scientists can then search these files via a web-based search engine for scientific data, called Science Search, that the Berkeley team is building. As a proof-of-concept, the team is working with staff at Berkeley Lab's Molecular Foundry, to demonstrate the concepts of Science Search on the images captured by the facility's instruments.

People who want to be inspired watch TED Talks, largely because of the stories they tell. These stories spark imagination and motivate audiences to think in new ways. TED Talks redefine knowledge and point to potential. Tufekci explains how we are routinely using computers to make subjective decisions like hiring new employees, releasing criminals from prison, and even identifying handwriting. Machine learning has made it possible for computers to make decisions about things we might not have disclosed – for example, about future likelihood of depression or pregnancy.

We study the problem of controlling linear time-invariant systems with known noisy dynamics and adversarially chosen quadratic losses. We present the first efficient online learning algorithms in this setting that guarantee $O(\sqrt{T})$ regret under mild assumptions, where $T$ is the time horizon. Our algorithms rely on a novel SDP relaxation for the steady-state distribution of the system. Crucially, and in contrast to previously proposed relaxations, the feasible solutions of our SDP all correspond to "strongly stable" policies that mix exponentially fast to a steady state.

With the recent success of embeddings in natural language processing, research has been conducted into applying similar methods to code analysis. Most works attempt to process the code directly or use a syntactic tree representation, treating it like sentences written in a natural language. However, none of the existing methods are sufficient to comprehend program semantics robustly, due to structural features such as function calls, branching, and interchangeable order of statements. In this paper, we propose a novel processing technique to learn code semantics, and apply it to a variety of program analysis tasks. In particular, we stipulate that a robust distributional hypothesis of code applies to both human- and machine-generated programs. Following this hypothesis, we define an embedding space, inst2vec, based on an Intermediate Representation (IR) of the code that is independent of the source programming language. We provide a novel definition of contextual flow for this IR, leveraging both the underlying data- and control-flow of the program. We then analyze the embeddings qualitatively using analogies and clustering, and evaluate the learned representation on three different high-level tasks. We show that with a single RNN architecture and pre-trained fixed embeddings, inst2vec outperforms specialized approaches for performance prediction (compute device mapping, optimal thread coarsening); and algorithm classification from raw code (104 classes), where we set a new state-of-the-art.